JP2006068489A - Interactive pet robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a novel interactive pet robot capable of proceeding conversation even if there is restriction on the number of utterance phrases and the number of response phrases. <P>SOLUTION: The interactive pet robot has a control part 9 for performing processing according to description of a program read out from a storage part. The program defines proceeding procedures of the conversation with a user using branches, in which a content of a question, a plurality of choices of answers to the question, and an action corresponding to each of the choices are associated. If the content of the question is outputted by sound, a plurality of choices associated with the question are displayed on a display 3 as propositions for the answer to the question. If a whole sentence of one of the choices displayed on the display 3 is uttered by the user, speech recognition is performed so as to identify one choice corresponding to utterance by the user, and to decide the action corresponding to the choice as the next reaction performed by the interactive pet robot corresponding to the answer. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an interactive pet robot, and more particularly to a pet robot suitable for language learning.

Conventionally, an interactive pet robot that communicates with a user is known. As one type, an interactive pet robot that is used for language learning such as English conversation is also known. Patent Document 1 discloses a language training robot that includes a voice recognition engine and translates a spoken native language into a foreign language and speaks. This robot has an external shape imitating the form of a duck or penguin, and a display screen is provided on its head. On this display screen, facial expressions (eyes, nose, mouth) are displayed, and the displayed facial expression changes according to the utterance of the translated language.
JP 2002-351305 A

  By the way, in such an interactive pet robot, how to promote the conversation with the user without delaying is an important point in improving the product appeal. To that end, it is conceivable to increase the degree of freedom of conversation by increasing the number of utterance phrases uttered by the robot and the number of standby phrases that can be heard by the robot. However, in this case, a high-speed microcomputer is required to recognize the user's utterance, and the description of the program becomes enormous, which makes it extremely difficult to commercialize at a low price range. On the other hand, in the conventional relatively inexpensive product, the user's utterance is often not recognized due to the restriction on the number of phrases described above, and the conversation tends to be delayed. As a result, there was a problem in the direction of communication between the interactive pet robot and the user.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a novel interactive pet robot that can promote conversation even if there are restrictions on the number of utterance phrases and the number of standby phrases. .

  Another object of the present invention is to enhance the sense of communication with the interactive pet robot and maintain the user's willingness to learn.

  In order to solve such a problem, a first invention includes a character-type robot body, a microphone attached to the robot body, a speaker attached to the robot body, a display attached to the robot body, and a program. Provided is an interactive pet robot having a storage unit stored therein and a control unit that is built in the robot body and performs processing according to the description of a program read from the storage unit. Here, the program defines the procedure for the conversation with the user with branching by the user's selection. In this program, the contents of a certain question, a plurality of options that are responses to the full text of the question, and actions for each of the options are described in association with each other. In addition, when the content of the question is output by voice from the speaker, the control unit performs a first process of displaying a plurality of options associated with the question on the display as a candidate for a response to the question. The control unit performs speech recognition using the user's full-text utterance obtained from the microphone as an input when any of a plurality of options displayed on the display is uttered by the user. A second process is performed in which an option corresponding to the option is specified, and an action corresponding to the option is determined as a response to be taken next by the interactive pet robot in response to the response.

  Here, in the first processing in the first invention, it is preferable that the control unit enumerates and displays the full text of a plurality of options together with the contents of the question. Further, in the first process, the control unit may emphasize and display the stress portion of inflection among the full text of the plurality of options.

  In the first invention, it is preferable that the action is to output an utterance as a response to the reply from the speaker.

  In the first invention, the control unit is preferably housed in the body of the robot body where the largest internal space is secured in the robot body, and the display is preferably mounted on the front of the body of the robot body.

  In the first invention, the control unit causes the display to display a plurality of words spelled and illustrations in order at regular intervals, and performs a third process of outputting the pronunciation of the words displayed in order from the speaker. You may go.

  In the first invention, the action may further include displaying the content of the utterance as a response to the response on the display.

  In the first process according to the first aspect of the invention, the control unit performs a plurality of options by performing voice recognition using a user's full-text utterance obtained from a microphone as input or operating a switch connected to the control unit. The fourth process of outputting at least one of the full texts as a model voice from the speaker may be performed.

  In the fourth process according to the first aspect, the control unit may emphasize the model voice and display it on the display among the full text of the plurality of options.

  On the other hand, the second invention is a character-type robot body, a microphone attached to the robot body, a presentation means attached to the robot body and presenting information to the user, a storage unit storing a program, Provided is an interactive pet robot having a control unit that is built in a robot body and performs processing according to a description of a program read from a storage unit. Here, the program defines the procedure for the conversation with the user with branching by the user's selection. In this program, the contents of a certain question, a plurality of options that are responses to the full text of the question, and actions for each of the options are described in association with each other. In addition, when the content of the question is presented to the user via the presenting means, the control unit presents a plurality of options associated with the question to the user via the presenting means in full text as a candidate for a response to the question. Process 1 is performed. The control unit performs speech recognition using the user's full-text utterance obtained from the microphone as an input when any of a plurality of options displayed on the display is uttered by the user. A second process is performed in which an option corresponding to the option is specified, and an action corresponding to the option is determined as a response to be taken next by the interactive pet robot in response to the response.

  Here, in the second invention, the presenting means is preferably at least one of a speaker and a display.

  According to the present invention, a plurality of options that are candidates for responses to questions of the interactive pet robot are presented to the user in full text. The user may select an appropriate one from a plurality of options presented in the full text and speak the full text as it is. On the other hand, for a system that performs speech recognition, since the user utters only the number of choices, the recognition can be easily performed without requiring a high-speed microcomputer. As a result, conversation can be promoted without causing the user to feel stress even with a relatively inexpensive product that has to be constrained in the number of utterance phrases and standby phrases. In addition, since information necessary for answering the question is presented by the pet robot, the user can proceed with the conversation while facing the pet robot. Thereby, a feeling of communication with a pet robot can be improved and a user's willingness to learn can be maintained effectively.

  FIG. 1 is a front view of an interactive pet robot according to the present embodiment, and FIG. 2 is a side view thereof. The interactive pet robot 1 includes a character-type robot body 2 and a display 3 attached to the robot body 2. The robot body 2 made of plastic has an appearance shape that is cute and deformed as an example of a character. The robot body 2 is mainly composed of a head, a torso and limbs, and a space for storing an electronic device is provided therein. The body 2a of the robot body 2 has a largest internal space in the robot body 2 so as to accommodate a control unit described later. The display 3 is a well-known small display such as a liquid crystal or an organic EL, for example, and is attached to the front surface of the body 2 a in the robot body 2. The reason why the display 3 is arranged in front is to ensure visibility in a state where the user faces the character, and the reason why this is arranged in the trunk 2a is that the control unit housed in the trunk 2a This is to shorten the connection distance of.

  FIG. 3 is a control block diagram of the interactive pet robot 1. The control system of the interactive pet robot 1 includes the display 3, the various sensors 4 to 7, the speaker 8, the control unit 9, and the various storage units 12 to 14. The microphone 4 is attached to an appropriate part of the robot body 2 and detects the user's utterance. The interpersonal sensor 5 is attached to an appropriate part of the robot body 2 and detects whether or not there is a person around the interactive pet robot 1. The forehead touch sensor 6 is attached to the forehead of the head of the robot body 2 and detects whether or not the user has touched the forehead. The four switches 7 are mounted at four positions of the right hand, the left hand, the right foot, and the left foot of the robot body 2, and detect whether or not the user has touched these parts. These sensors 4 to 7 enable the system side to recognize an action (for example, touching the forehead, grasping a limb) as a communication performed by the user.

  The controller 9 includes a voice recognition unit 10 and a main control unit 11. The voice recognition unit 10 performs voice recognition using the user's utterance obtained from the microphone 4 as an input, and outputs the recognition result to the main control unit 11. Since the voice recognition process using the user's utterance as an input is well known, the description thereof is omitted here. The main control unit 11 mainly controls the recognition result from the voice recognition unit 10 and controls the display 3 and the speaker 8 after appropriately adding detection signals from the sensors 5 to 7 to this. A mask ROM 12 connected to the main control unit 11 stores a program (main program) that defines a procedure for a conversation with the user. The SRAM connected to the main control unit 11 is a working memory for the main control unit 11. The external ROM connected to both the voice recognition unit 10 and the main control unit 11 is a storage unit in an additional cartridge that can be attached to and detached from the robot body 2. The external ROM stores a program (extended program) that provides an extended function of the main program described above.

  FIG. 4 is a flowchart showing an example of a procedure for progressing conversation defined by the program. This program defines the procedure of conversation with the user with branching by the user's selection. The main control unit 11 performs processing according to the following procedure in accordance with the description of the program read from the mask ROM 12.

  First, in the standby state of Step 1, three options “Let's talk.”, “Wake up.” And “Play.” Are displayed on the display 3. The user selects an arbitrary one from these options and utters the full text. The voice recognition unit 10 performs voice recognition using the user's full-text utterance obtained from the microphone 4 as input, and outputs the recognition result to the main control unit 11. The main control unit 11 specifies one option (“Let's talk.”, “Wake up.” Or “Play.”) Corresponding to the user's utterance based on the recognition result from the speech recognition unit 10. Then, the main control unit 11 determines an action corresponding to the specified option as a reaction that the pet robot 1 should take next. For example, if the specified option is “Let's talk.” Or “Wake up.”, The utterance in step 2 is output from the speaker 8, and if “Play.”, The utterance in step 10 is audio. It seems to be output.

  The subsequent steps 2 to 5 form a series of flows such as a question from the pet robot 1, a response from the user, and a reaction of the pet robot 1 to this, and are the core parts of conversation (communication). In this regard, the program describes the contents of a certain question, a plurality of options that are responses to the full text of the question, and a series of actions that are actions for each of the options in association with each other. Specifically, for the question “Hi. How are you? My friend.”, Three options (full text) “I'm fine.”, “Great!” And “Not so good.” ing. In addition, the choices “I'm fine.” And “Great!” Are associated with random actions by voice output “That's great!” And “That's good, but I have a headache.” The option “Not so good.” Is associated with an action by voice output “Oh, really? I'm sorry to hear that. I hope you will feel better soon.”.

  In accordance with the description of such a program, in step 2, the main control unit 11 controls the speaker 8, whereby a question “Hi. How are you? My friend.” Is output from the speaker 8 by voice. In the subsequent step 3, the main control unit 11 controls the display 3, so that the full text of the three options associated with the question is displayed on the display 3 as a candidate for a response to the question.

  FIG. 5 is a diagram illustrating a display example of options that are candidates for a response to a question. In this display example, the content of the question that was output as voice, that is, “Hi. How are you? My friend.” Is displayed at the top, and “I'm fine.”, “Great!” Three options “Not so good.” Are listed and displayed in full text. Thereby, since the user can grasp at a glance the correspondence between the contents of the questions and the contents of the options, the learning efficiency can be improved particularly in the case of language learning. Further, in this display example, the stress portion of the inflection in the full text of the options is highlighted and displayed by changing the font color and size, etc., in order to attract the user's attention and further improve the learning efficiency. The user selects an arbitrary one from these options and utters the full text.

  At this time, the pet robot 1 may cause each of the options spoken by the user to be pronounced as a model voice. For example, in the state shown in FIG. 5, the user performs a predetermined operation on the switch 7. Then, in response to this operation, the main control unit 11 controls the speaker 8, and at least one of the options is output from the speaker 8 as a model voice. When all the output of the model voice is completed, the state returns to the state shown in FIG.

  In addition to the operation of the switch 11, the start of the model voice may be, for example, voice recognition from the microphone 4 for the word “model!” From the user. Further, after the “example!”, A number “1” indicating a choice is recognized by the microphone 4 so that only the third of the choices is outputted as an example voice from the speaker 8.

  Furthermore, when the pet robot 1 generates this model voice, it may be displayed on the display 3 to highlight which option is pronounced as the model voice. For example, one option (for example, “I'm fine.”) Currently displayed on the display 3 is pronounced as a model voice from the speaker 8. At this time, the main control unit 11 controls the display 3 to highlight the character (or option) corresponding to this one option. Note that highlighting methods include changing the font color and size of characters (or options), underlining, changing the font, and changing to bold characters.

  The voice recognition unit 10 performs voice recognition using the user's full-text utterance obtained from the microphone 4 as an input, and outputs the recognition result to the main control unit 11. The main control unit 11 identifies one option (“I'm fine.”, “Great!” Or “Not so good.”) Corresponding to the user's utterance based on the recognition result from the speech recognition unit 10. To do. Then, the main control unit 11 determines an action corresponding to the specified option as a reaction that the pet robot 1 should take next. If the specified option is “I'm fine.” Or “Great!”, Go to Step 4 and select either “That's great” or “That's good, but I have a headache.” This is selected at random, and this is output as audio from the speaker 8. If the specified option is “Not so good.”, Go to step 5 and hear the speech “Oh, really? I'm sorry to hear that. I hope you will feel better soon.” Is output.

  If “That's great” is output as a voice in step 4, the process proceeds to step 10 and the question “What do you want to do?” Is output as a voice, while “That's good, but I have” in step 4. If "a headache." is output as a voice, in step 6, "Please touch my head gently." is output as a voice, and then the user waits for an action (step 7). When the user touches the forehead of the pet robot 1, this action is detected by the forehead touch sensor 6. Accordingly, in Step 8, as a reaction of the interactive pet robot 1, “Thank you my friend. “I feel much better now.” Is output, and then the process proceeds to step 10 described above. On the other hand, if the user touches other than the forehead and a sensor other than the forehead touch sensor 6 detects this, In step 9, the voice of “Please touch my head. Try again.” Is output as a response of the interactive pet robot 1 and the process returns to the waiting state of step 7. As a result, the user touches the forehead. Until then, the loop of steps 7 and 9 is repeated.

If "What do you want do?" Is output as a voice in step 10, the process proceeds to step 11 to be an option for each section. For example, when a favorite fruit section is selected, the conversation is related to the interactive pet robot 1 and the user's favorite food.

  In the above-described conversation progress procedure, when the user utters “See you later.”, It is set to forcibly return to the standby state in Step 1. Further, when the user utters “Go back.”, It is set to return to step 10 forcibly.

  Further, in the conversation proceeding procedure as described above, when the pet robot 1 outputs a question or a response of a response by voice, the utterance content may be displayed on the display 3. For example, in step 2, when the content “Hi How are you? My friend.” Is output from the speaker 8 as audio, the main control unit 11 controls the display 3 so that “Hi How are you? My friend.” The question "How are you? My friend."

  In addition to the above, the additional cartridge with built-in external ROM 14 assumes that the robot is an interviewer in order to further improve the English learning ability, and the “Interview of Eiken Level 3” that displays the question card on the display. There is a virtual place. There is also a “flash card game” that can learn the pronunciation and spelling of words. In the latter case, the main control unit 11 causes the display 3 to display the spelling of the word and the illustration (for example, the spelling of the book and the illustration of the book) in order at regular intervals, and also displays the pronunciation of the displayed word from the speaker 8. Output audio. The robot responds with “Good.” If the user pronounces correctly, “Try again” if not. In addition, various series are available, such as those based on practical conversations assuming “shopping abroad”.

  As described above, according to the present embodiment, a plurality of options that are candidates for a response to the question of the interactive pet robot 1 are presented to the user in full text. The user may select an appropriate one from the plurality of presented options and utter the full text as it is. On the other hand, for the voice recognition unit 10, since the number of variations that the user utters is limited to the number of options, the recognition can be performed easily and at high speed without requiring a high-speed microcomputer. As a result, even if it is a relatively inexpensive product that must place restrictions on the number of spoken phrases and standby phrases, it is possible to promote conversation without making the user feel stressed by presenting the full text of the options. .

  Further, according to the present embodiment, information necessary for answering a question is presented from the interactive pet robot 1, so that the user can refer to other information while facing the pet robot 1. You can have a conversation without it. Thereby, a feeling of communication with the pet robot 1 can be enhanced, and the user's willingness to learn can be effectively maintained.

  In the above-described embodiment, the operation of the interactive pet robot 1 has been described by taking English conversation as an example. However, the present invention is not limited to this, and can be widely applied to conversation learning in each language. is there.

  Further, the action to be taken by the interactive pet robot 1 is not limited to the sound output from the speaker 8. For example, if the limbs of the interactive pet robot 1 are movable, they may be operated as actions, and if a light emitter is attached, they may be flashed.

  Further, in the above-described embodiment, options that are candidates for answering the question are displayed in full text on the display 3, but this can also be presented to the user by voice output of the full text from the speaker 8. However, from the viewpoint of promoting conversation, it is preferable to display on the display 3 that does not require time for speaking, rather than the interactive pet robot 1 speaking the choices for the question each time.

Appearance front view of interactive pet robot Appearance side view of interactive pet robot Control block diagram of interactive pet robot Flow chart showing the progress of conversation The figure which shows the example of a display of the choice which is the candidate of the reply to a question

Explanation of symbols

1 Interactive Pet Robot 2 Robot Body 2a Body 3 Display 4 Microphone 5 Human Sensor 6 Forehead Touch Sensor 7 Switch 8 Speaker 9 Control Unit 10 Voice Recognition Unit 11 Main Control Unit 12 Mask ROM
13 SRAM
14 External ROM

Claims (11)

In interactive pet robot,
A character-type robot body,
A microphone attached to the robot body;
A speaker attached to the robot body;
A display attached to the robot body;
The procedure of the conversation with the user is defined with a branch by the user's selection, and the contents of a question, a plurality of options that are responses to the full text of the question, and actions for each of the options are associated with each other A storage unit storing the described program;
A control unit built in the robot body and performing processing according to the description of the program read from the storage unit;
The controller is
When the content of the question is output by voice from the speaker, as a candidate for a response to the question, a first process is performed to display the plurality of options associated with the question on the display in full text,
When any of the plurality of options displayed on the display is uttered by the user, the user's utterance is input by performing speech recognition using the user's full-text utterance obtained from the microphone as an input. And a second process of determining the action corresponding to the one option as the next response to be taken by the interactive pet robot with respect to the response. Interactive pet robot.   2. The interactive pet robot according to claim 1, wherein in the first process, the control unit enumerates and displays the full text of the plurality of options together with the contents of the question on the display.   2. The interactive pet robot according to claim 1, wherein, in the first process, the control unit emphasizes a stress portion of inflection among all the texts of the plurality of options and displays the stress on the display. 3. .   2. The interactive pet robot according to claim 1, wherein the action is to output an utterance as a response to the response by voice from the speaker. The control unit is housed in the body of the robot body in which the largest internal space is secured in the robot body,
The interactive pet robot according to claim 1, wherein the display is mounted on a front surface of a body portion of the robot body.   The control unit displays a spelling and illustrations of a plurality of words on the display in order at regular intervals, and performs a third process of outputting the pronunciation of the words displayed in order by voice from the speaker. The interactive pet robot according to claim 1.   The interactive pet robot according to claim 4, wherein the action further includes causing the display to display a content of an utterance as a response to the response.   In the first process, the control unit performs voice recognition using a user's full-text utterance obtained from the microphone as an input, or operates a switch connected to the control unit, thereby the plurality of options. 2. The interactive pet robot according to claim 1, wherein at least one of the full texts is output as voice from the speaker as a model voice.   9. The interactive pet robot according to claim 8, wherein in the first process, the control unit highlights the model voice among the whole sentences of the plurality of options and displays the emphasized voice on the display. In interactive pet robot,
A character-type robot body,
A microphone attached to the robot body;
Presenting means mounted on the robot body and presenting information to a user;
The procedure of the conversation with the user is defined with a branch by the user's selection, and the contents of a question, a plurality of options that are responses to the full text of the question, and actions for each of the options are associated with each other A storage unit storing the described program;
A control unit built in the robot body and performing processing according to the description of the program read from the storage unit;
The controller is
When the content of the question is presented to the user via the presenting means, the plurality of options associated with the question are presented to the user via the presenting means in full text as candidates for a response to the question. 1 process,
When any of the plurality of options displayed on the display is uttered by the user, the user's utterance is input by performing speech recognition using the user's full-text utterance obtained from the microphone as an input. And a second process of determining the action corresponding to the one option as the next response to be taken by the interactive pet robot with respect to the response. Interactive pet robot.   The interactive pet robot according to claim 10, wherein the presenting means is at least one of a speaker and a display.