JP2001209644A - Information processor, information processing method and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily provide the message of an expression appropriate for an opposite party. SOLUTION: The speech of a user is voice-recognized in a voice recognition part 1, language-processed in a language processing part 2 and then supplied to an intention understanding part 3. In the intention understanding part 3, the opposite party to transmit the message and contents desired to be transmitted are recognized. In a sentence generation part 6, based on the recognized result, the message of the contents desired to be transmitted by the expression corresponding to intimacy with the opposite party to transmit the message is generated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing apparatus, an information processing method, and a recording medium, and more particularly to, for example, an information processing apparatus capable of transmitting a message of an appropriate expression to a partner. The present invention relates to an information processing method, and a recording medium.

[0002]

2. Description of the Related Art Hitherto, as a means for transmitting a message from one person to another person, oral, telephone, letter,
Furthermore, in recent years when the Internet has rapidly spread, there is e-mail and the like.

[0003]

When the message is transmitted to a partner such as a brother, friend, junior, etc., the user does not need to pay much attention to the expression of the message.

[0004] However, for example, when a message is transmitted to a boss, a senior, a teacher, or the like, it is often troublesome to create a message while paying attention to the expression.

The present invention has been made in view of such a situation, and is intended to easily create and transmit a message having an appropriate expression to a partner.

[0006]

SUMMARY OF THE INVENTION An information processing apparatus according to the present invention analyzes an input from a user and analyzes at least a partner to transmit a message and contents to be transmitted, and a partner to transmit a message. And a message generating means for generating a message of the content to be transmitted in an expression according to the intimacy with the message.

[0007] The information processing apparatus may further include a transmitting means for transmitting the message generated by the message generating means to a partner transmitting the message.

[0008] The transmission means can output the message as a synthesized sound.

The transmission means can transmit the message via a network.

[0010] The information processing apparatus of the present invention can be a robot.

The message generating means can generate a message which does not overlap with a message generated in the past.

The information processing apparatus according to the present invention may further include a storage unit for storing the history of the message generated by the message generation unit. In this case, the message generation unit refers to the storage contents of the storage unit. By doing
It is possible to generate a message that does not overlap with a message generated in the past.

The message generating means can generate a message that does not overlap with the message by replacing one or more words of the message generated in the past.

[0014] The information processing apparatus of the present invention may further include voice recognition means for recognizing the voice of the user. In this case, the analysis means can analyze the voice recognition result by the voice recognition means.

An information processing method according to the present invention analyzes an input from a user and analyzes at least a partner to which a message is to be transmitted and contents to be transmitted, and a method according to the degree of intimacy with the partner to which the message is to be transmitted. And a message generating step of generating a message of the content to be transmitted in the above expression.

The recording medium according to the present invention analyzes an input from a user, analyzes at least a partner to which a message is to be transmitted, and a content to be transmitted, and responds to an intimacy with the partner to which the message is transmitted. A message generating step of generating a message of the content to be transmitted by the expression.

In the information processing apparatus, the information processing method, and the recording medium of the present invention, an input from a user is analyzed, and at least a partner to which a message is to be transmitted and contents to be transmitted are recognized. Then, a message having contents to be transmitted is generated by an expression according to the intimacy with the partner to which the message is transmitted.

[0018]

FIG. 1 shows an example of the external configuration of a robot according to an embodiment of the present invention.

This robot transmits a message from a user to a predetermined partner.

That is, in the present embodiment, the robot has a dog shape, and leg units 3A, 3B, 3C, 3D are connected to the front, rear, left and right of the body unit 2, respectively. The head unit 4 and the tail unit 5 are connected to the front end and the rear end of the unit 2 respectively.

The tail unit 5 is drawn out from a base unit 5B provided on the upper surface of the body unit 2 so as to bend or swing with two degrees of freedom.

The joints of the leg units 3A to 3D, the joints of the leg units 3A to 3D and the body unit 2, the joints of the head unit 4 and the body unit 2, and the tail unit 5 An actuator (not shown) is provided at a connection portion of the body unit 2 and the like, whereby each connection portion rotates with a predetermined degree of freedom, so that the robot can take an actual dog. You can take actions similar to actions.

In this embodiment, as shown in FIG. 2A, the robot recognizes the utterance of the user by voice, analyzes the voice recognition result, and wants to transmit the message to the other party. It is designed to recognize the contents. Further, the robot generates a message of the content to be transmitted in an expression according to the intimacy (friendliness) between the user and the user who transmits the message, and for example, as shown in FIG. Go to the place where there is, and transmit the generated message by synthetic sound.

Further, the robot is provided with a communication function, so that the generated message can be transmitted by e-mail or the like as necessary.

Next, FIG. 3 shows an example of the electrical configuration of the robot shown in FIG.

A speech uttered by the user is input to the speech recognition unit 1. The speech recognition unit 1 performs speech recognition of the inputted speech, and outputs a text (phoneme information) as the speech recognition result. ) Is output to the language processing unit 2.

The language processing section 2 language-processes the speech recognition result output from the speech recognition section 1 and uses the words, syntax and meaning information included in the speech recognition result as a language processing result as an intention understanding section 3. Output to

The intention understanding unit 3 uses the language processing result of the language processing unit 2 to send a message to a user (hereinafter, referred to as a user).
Understand the contents (intent) you want to convey, as appropriate. Further, the intention understanding unit 3 generates sentence generation control information as control information for generating a message of the content to be conveyed in an expression corresponding to the familiarity between the understood communication partner and the user, and the sentence generation unit 6 Output to

The sentence generation unit 6 generates a message according to the sentence generation control information from the intention understanding unit 3 and outputs the message to the speech synthesis unit 7 or the communication unit 9.

That is, when the communication partner is in a place where the robot can move, the sentence generation unit 6 outputs the generated message to the speech synthesis unit 7. Also,
In this case, the intention understanding unit 4 supplies the control unit 4 with location information indicating the location of the communication partner. When receiving the location information, the control unit 4 controls the driving unit 5 to move the robot to the location represented by the location information. Drive unit 5
Drives the actuator in accordance with the control from the control unit 4, whereby the robot moves to the location represented by the location information, that is, the location where the communication partner exists. After that, the speech synthesis unit 7 generates a synthesized sound corresponding to the message from the sentence generation unit 6 and outputs it from the speaker 8.

On the other hand, when the communication partner is in a place where it is difficult for the robot to move, the sentence generation unit 6 outputs the generated message to the communication unit 9. Communication unit 9
Has a function similar to that of a mobile phone capable of transmitting and receiving e-mail, for example, and converts a message from the sentence generation unit 9 into an e-mail and sends the message to a communication partner via a predetermined network such as the Internet. Send to e-mail address.

The communication section 9 can be configured to be connectable to a connector or a contact point connected to a predetermined network such as a public network. In this case, for example, the robot moves to the position where the connectors and contacts are installed.
It is possible to autonomously move, connect the communication unit 9 and a connector or the like, and transmit an e-mail.

Next, FIG. 4 shows an example of the configuration of the voice recognition unit 1 of FIG.

The user's utterance is input to a microphone (microphone) 11, where the utterance is converted into an audio signal as an electric signal. This audio signal is supplied to an A / D (Analog Digital) converter 12.
The A / D converter 12 samples and quantizes an audio signal, which is an analog signal from the microphone 11, and converts it into audio data, which is a digital signal. This audio data is supplied to the feature extraction unit 13.

The feature extraction unit 13 converts the audio data from the A / D conversion unit 12 into, for example, a spectrum, a linear prediction coefficient, a cepstrum coefficient,
Line spectrum pair, MFCC (Mel Frequency Cepstrum Co.)
A characteristic parameter such as efficient is extracted and supplied to the matching unit 14.

The matching unit 14 refers to the acoustic model database 15, the dictionary database 16, and the grammar database 17 as needed based on the feature parameters stored in the feature extraction unit 13 and, if necessary, inputs the speech input to the microphone 11. Recognize (input voice).

That is, the acoustic model database 15 stores acoustic models representing acoustic features such as individual phonemes and syllables in the language of the speech to be recognized. Here, as the acoustic model, for example, HMM (Hidden Mark
ov Model) can be used. The dictionary database 16 stores a word dictionary in which information about pronunciation is described for each word to be recognized. The grammar database 17 stores grammar rules that describe how the words registered in the word dictionary of the dictionary database 16 are linked (connected). Here, the grammar rules include, for example, context-free grammar (CFG) and H
PSG (Head-driven Phrase Structure Grammar) (head-driven phrase structure grammar), statistical word chain probability (N-gr
am) or the like.

The matching unit 14 stores the dictionary database 1
By connecting the acoustic models stored in the acoustic model database 15 by referring to the word dictionary of No. 6, an acoustic model (word model) of the word is formed. Further, the matching unit 14 connects several word models by referring to the grammar rules stored in the grammar database 17, and uses the thus connected word models based on the feature parameters, for example, , The HMM method or the like is used to recognize the voice input to the microphone 11.

The phoneme information as a result of the speech recognition by the matching unit 14 is output to the language processing unit 2, for example, as text.

FIG. 5 shows an example of the configuration of the language processing section 2 shown in FIG.

The text analysis unit 21 includes the speech recognition unit 1
The text as a speech recognition result output by the (matching unit 14 in FIG. 4) is input. The text analysis unit 21 refers to the dictionary database 23 and the grammar database for analysis 24 and converts the text. To analyze.

That is, the dictionary database 23 stores a word dictionary in which notation of each word, part of speech information necessary for applying the grammar for analysis, and the like are described. The analysis grammar database 24 stores analysis grammar rules in which restrictions on word chains and the like are described based on information on each word described in the word dictionary. Then, the text analysis unit 21 uses the word dictionary and the grammar rules for analysis to input the text (speech recognition result) therein.
And outputs the analysis result to the syntax / semantic analysis unit 22.

The syntax / semantic analysis unit 22 refers to the dictionary database 25, the analysis grammar database 26, and the command dictionary database 27, and
Based on the output, the syntax analysis of the speech recognition result by the speech recognition unit 1 and the understanding of the semantic content are performed. Then, the syntax / semantic analysis unit 22 adds information representing the concept and the meaning of each word constituting the speech recognition result to the speech recognition result by the speech recognition unit 1 as the processing result, and Output to the unit 3 (FIG. 3).

Here, the dictionary database 25 and the analysis grammar database 26 include, for example, the dictionary database 2
3 and the same information as those in the analysis grammar database 24 are stored. Further, the command dictionary database 27 stores a command dictionary in which words as commands given from the user to the robot are described. The syntax / semantic analysis unit 22 performs syntax analysis and semantic understanding using, for example, regular grammar, context-free grammar, HPSG, and statistical word chain probability.

Next, FIG. 6 shows an example of the configuration of the intention understanding unit 3 in FIG.

The keyword extracting unit 31 includes the language processing unit 2
The processing result (language processing result) of the (syntax / semantic analysis unit 22 in FIG. 5) is input. The keyword extraction unit 31 refers to the dictionary database 33 and the person database 34, and A word (string) serving as a keyword is extracted from the result of the second language processing.

That is, the dictionary database 33 stores a word dictionary in which words useful for understanding the intention of the utterance content of the user are registered together with the part of speech information.
In addition, the person database 34 stores, for example, person data as shown in FIG. 7 as information regarding a partner to whom the utterance content of the user is transmitted (a transmission partner).

Here, in the embodiment shown in FIG. 7, the person data includes a name for specifying a person to whom the utterance contents of the user are to be transmitted, an e-mail address of the person, a place where the person is located, and a name of the person. It is composed of intimacy with the user.

In the keyword extracting section 31, the language processing section 2
From the linguistic processing results of the above, words registered in the word dictionary of the dictionary database 33 are extracted as keywords,
It is supplied to the intention recognition unit 32. Further, in the keyword extraction unit 31, the name of the communication partner registered in the person database 34 is extracted from the language processing result of the language processing unit 2, and supplied to the intention recognition unit 32. When the name of the communication partner is extracted from the language processing result of the language processing unit 2, the keyword extraction unit 31 reads the e-mail address, location, and intimacy of the communication partner from the person database 34. These are also supplied to the intention recognition unit 32.

The intention recognizing unit 32 includes a keyword extracting unit 31
, Generates sentence generation control information as control information for causing the sentence generation unit 6 (FIG. 3) to generate a message, and supplies it to the sentence generation unit 6 (FIG. 3).

That is, the intention recognition unit 32 recognizes the intention of the utterance content of the user based on the keyword in the output of the keyword extraction unit 31. Then, the intention recognition unit 32
Supplies a command instructing generation of a message reflecting the intention together with the output of the keyword extraction unit 31 to the sentence generation unit 6 as sentence generation control information.

The user may utter a command (for example, “sit” or “hand”) given to the robot in addition to a message to be transmitted to someone. In the case of such a command, the language processing result as the command is supplied to the control unit 4 (FIG. 3) via the keyword extraction unit 31 and the intention recognition unit 32. In this case, the control unit 4 controls the driving unit 5 so as to perform an operation corresponding to the command, whereby the robot performs such an operation (takes an action).

When a message is output as a synthesized sound as described above, the robot needs to move to the location of the transmission partner that transmits the message. Therefore, when the message is output as a synthesized sound, the intention recognition unit 32 supplies the sentence generation control information to the sentence generation unit 6 and reads the sentence generation control information from the person database 34 included in the sentence generation control information. The issued location is output to the control unit 4. In this case, the control unit 4 controls the driving unit 5 so as to move to the place from the intention recognition unit 32, whereby the robot moves to the place where the transmission partner exists.

Next, FIG. 8 shows a configuration example of the sentence generation unit 6 of FIG.

The text generation unit 41 includes the intention understanding unit 3
The sentence generation control information output by the (intention recognition unit 32 in FIG. 6) is supplied, and the text generation unit 41
Sentence history storage unit 42, template database 43, generated grammar database 44, and dictionary database 45
Is generated as needed based on the sentence generation control information, and supplied to the speech synthesis unit 7 or the communication unit 9 while referring to as necessary.

That is, the sentence history storage unit 42 stores the history of messages generated by the text generation unit 41 (hereinafter, appropriately referred to as sentence history).

FIG. 9 shows the sentence history stored in the sentence history storage unit 42.

The sentence history includes an identification number, the number of times of use, and the date and time of use. The identification number is used to identify a template stored in the template database 43 used by the text generation unit 41 to generate a message. Represents a number. The usage count indicates the number of times the corresponding template has been used for generating a message by the text generation unit 41, and the use date and time indicates the latest date and time when the corresponding template has been used.

Note that the sentence history is stored in the text
Each time a template stored in the template database 43 is used, the text is updated by the text generator 41.

The template database 43 stores templates used by the text generator 41 for generating messages, and the generated grammar database 44 stores rules for utilizing words necessary for generating messages, and rules for word order. Grammar rules such as constraint information are stored. In addition, the dictionary database 45 stores a word dictionary in which part-of-speech information of each word, and information such as readings and accents are described. Then, the text generation unit 41 refers to these sentence histories, templates, grammar rules, and word dictionaries as necessary, and generates a message according to the sentence generation control information from the intention understanding unit 3.

That is, the text generation unit 41 searches for one or more templates that are included in the sentence generation control information and that reflect the intention of the utterance content of the user and are not used much (the ones that are used less frequently). And then, part 1
From the above-mentioned templates, an expression corresponding to the intimacy with the communication partner included in the sentence generation control information is selected. The number of times the template is used is recognized by referring to the number of times the sentence history (FIG. 9) is used.

Then, the text generation unit 41 determines whether a word extracted as a keyword included in the sentence generation control information or the dictionary database 4
In accordance with the grammatical rules stored in the generated grammar database 44, the words registered in the word dictionary of No. 5 are inserted as necessary while appropriately changing the inflected forms, etc. A text message reflecting the intention of the utterance content of the user is generated by the expression.
This message is output to the speech synthesis unit 7 or the communication unit 9 as described above.

When a message is output to the communication section 9 and transmitted to the transmission destination by e-mail, an e-mail address of the transmission destination, which is the destination of the e-mail, is required. Is included in the sentence generation control information, and is supplied from the text generation unit 41 to the communication unit 9 together with the generated message.

As described above, in the text generation unit 41,
It generates a message in an expression according to the intimacy with the transmission partner included in the sentence generation control information,
In order to generate such a message, templates of various expressions are stored in the template database 43 for messages having the same contents.

FIG. 10 shows templates stored in the template database 43.

This template is for a message for guiding the holding of an event. FIG. 10A shows a template used when a person who shows respect such as a senior or a teacher is a communication partner. And FIG.
(B) shows a template used when a close person such as a friend is a communication partner.

Therefore, the template shown in FIG.
It is a polite expression, and the template of FIG. 10 (B) is a somewhat rough expression.

In the template shown in FIG.
The words corresponding to o, Date, Place, What, and Topic are inserted by the text generator 41. Here, To represents a transmission partner, and Date represents a date and time when the event is held. “Place” indicates a place where the event is held, and “What” indicates the event. Furthermore, Topic represents some topic. Then, in To, the name of the communication partner included in the sentence generation control information is inserted. Date, Place, W
In hat, a word as a corresponding keyword included in the sentence generation control information is inserted. In addition, Topic has
Appropriate words among words registered in the word dictionary in the dictionary database 45 are inserted in advance.
However, the words arranged in the Topic can be replaced by the text generation unit 41 as necessary, as described later.

Each template in the template database 43 is added with a tag indicating the content of the template for the message, and the text generation unit 41 refers to the tag. By doing so, a search that reflects the intention of the utterance content of the user (described as “Subject” in FIG. 13 described later) included in the sentence generation control information is searched from the template of the template database 43.

As a method of generating a message in the text generation unit 41, for example, a method based on a case structure or the like can be adopted in addition to a method using a template.

FIG. 11 shows an example of the configuration of the speech synthesizer 7 in FIG.

The text analysis section 51 is supplied with a text message output by the sentence generation section 6 (the text generation section 41 in FIG. 8).
1 analyzes a text as a message supplied thereto while referring to the dictionary database 54 and the analysis grammar database 55.

That is, the dictionary database 54 stores a word dictionary in which part of speech information of words used in a message, and information such as reading and accent are described. The analysis grammar database 55 stores a dictionary. For words described in the word dictionary of the database 54, grammatical rules for analysis such as restrictions on word chains are stored. Then, the text analysis unit 51 analyzes the message from the sentence generation unit 6 such as morphological analysis and syntax analysis based on the word dictionary and the grammatical rules for analysis. Extract information required for synthesis. Here, the information necessary for the rule speech synthesis includes, for example, information for controlling the position of a pause, accent and intonation, other prosody information, and phoneme information such as pronunciation of each word.

The information obtained by the text analysis section 51 is supplied to a rule synthesis section 52, which uses a phoneme segment database 56 to generate a synthesized speech corresponding to the message generated by the sentence generation section 6. Is generated.

That is, the phoneme unit database 56 stores phoneme unit data in the form of, for example, CV (Consonant, Vowel), VCV, CVC, and the like.
Based on the information from the text analysis unit 51, necessary phoneme segment data is connected, and a pause, an accent, an intonation, and the like are appropriately added, thereby synthesizing the response sentence generated by the sentence generation unit 6. Generate audio data for the sound.

This audio data is stored in a digital / analog (D / A)
g) The signal is supplied to the conversion unit 53, where it is converted into an audio signal as an analog signal. This audio signal is transmitted to the speaker 8
(FIG. 3), whereby a synthesized sound corresponding to the message generated by the sentence generation unit 6 is output.

Next, with reference to the flowchart of FIG. 12, a description will be given of the processing in the case where the robot of FIG. 3 transmits a message intended by the user to a predetermined partner.

When the user speaks, the user's voice is input to the voice recognition unit 1, and the voice recognition unit 1 recognizes the user's voice in step S1. Here, in this case, it is assumed that the user's utterance includes at least a matter to be conveyed to somebody and a person who conveys the matter (communication partner).

The speech recognition result obtained by performing the speech recognition in step S1 is supplied to the language processing unit 2, and the language processing unit 2 language-processes the speech recognition result in step S2, Is supplied to the intention understanding unit 3.

In step S3, the intention understanding unit 3
Based on the language processing result from the language processing unit 2, the above-described sentence generation control information is generated and supplied to the sentence generation unit 6. In addition, when outputting a message as a synthetic sound,
As described above, the intention understanding unit 3 further outputs to the control unit 4 information on the location of the communication partner who transmits the message. Thereby, the robot starts moving to the location of the communication partner.

Here, in order for the robot to move to a predetermined destination, it is necessary for the robot to recognize its own current location. For example, this recognition of the current location is performed by recognizing an image of the surroundings of the robot. , Based on the recognition result. In addition, the current location is recognized by GPS (Global Posi
It is also possible to carry out this using a method such as an arresting system.

After the process in step S3, the process proceeds to step S4, in which the sentence generation unit 6 determines the contents of the user's utterance in an expression corresponding to the intimacy with the communication partner in accordance with the sentence generation control information from the intention understanding unit 3. Generates a message reflecting the
Output to the voice synthesizing unit 7 or the communication unit 9. Note that the sentence generation unit 6 outputs the message to the communication unit 9 by
The e-mail address of the communication partner is output to the communication unit 9 together with the message.

Then, the process proceeds to step S5, in which the message generated by the sentence generating unit 6 is transmitted to the transmission partner in the voice synthesizing unit 7 or the communication unit 9, and the process is terminated.

That is, when the message output from the sentence generating unit 6 is supplied to the speech synthesizing unit 7, as described above, after the robot starts moving to the location of the communication partner,
After arriving at the place, speech synthesis is performed by the speech synthesis unit 7, and a synthesized sound corresponding to the message output by the sentence generation unit 6 is output from the speaker 8.

The message output from the sentence generation unit 6 is transmitted to the communication unit 9 together with the e-mail address of the communication partner.
In this case, the communication unit 9 creates an e-mail with the message from the sentence generation unit 6 as the content (text) with the e-mail address as the destination, and transmits the e-mail via the network. When an e-mail is transmitted wirelessly, if an antenna is required, the tail unit 5 (FIG. 1) can be used as the antenna.

According to the above processing, for example, the user
Suppose that "Please give Sony-Taro a guide about the exhibition at the International Exhibition Center on December 24." In the speech recognition unit 1, for example, FIG.
A speech recognition result as shown in (A) is obtained, and further, the speech recognition result is subjected to language processing in the language processing unit 2 and processed in the intention understanding unit 3, so that, for example,
Sentence generation control information including information as shown in FIG. 13B is generated.

In this case, the content of the user's utterance guides the holding of the event, and the sentence generation unit 6 generates a template of such content and transmits the template to the communication partner Sony-Taro. For example, a message as shown in FIG. 13C is generated using an expression corresponding to the intimacy level. Here, in FIG. 13 (C), the part enclosed by brackets [] is a word (column) extracted from the user's utterance, which is inserted into the template.

The message generated as described above is output as a synthesized sound or transmitted as an electronic mail.

As described above, by analyzing the utterance of the user, it is possible to recognize the partner to whom the message is to be transmitted and the content to be transmitted, and to transmit the content to be transmitted in an expression according to the familiarity with the partner to which the message is to be transmitted. Is generated, a message with an appropriate expression can be easily created (minimized the burden on the user) and transmitted to the other party. Therefore, the robot of FIG.
For example, in a company, it can be used to transmit a report from a subordinate to a supervisor, etc. In this case, the robot
For a boss, it can function as an electronic secretary.

Next, with reference to the flowchart in FIG. 14, the processing performed by the sentence generation unit 6 in step S4 in FIG. 12 will be further described.

The text generator 41 of the sentence generator 6 (FIG. 8)
Receives the sentence generation control information from the intention understanding unit 3, selects a message template having the content indicated by the sentence generation control information and having a small number of uses in step S11.

That is, the text generation unit 41 generates a template that reflects the intention of the user's utterance content included in the sentence generation control information and is rarely used.
Search above, and from the one or more templates,
An expression corresponding to the intimacy with the communication partner included in the sentence generation control information is selected. Note that the number of times the template is used is recognized by referring to the sentence history (FIG. 9) as described above.

Then, proceeding to step S12, the text generation section 41 determines whether or not the template selected in step S11 (hereinafter, appropriately referred to as a selected template) has been used recently. Whether the selected template has been used recently is determined by referring to the use date and time of the sentence history (FIG. 9).

If it is determined in step S12 that the selected template has been used recently, the process returns to step S11 and the template is selected again. However, in this case, redoing of template selection is performed for templates other than those already selected as the selected template among the templates stored in the template database 43.

On the other hand, if it is determined in step S12 that the selected template has not been used recently, the process proceeds to step S13, where the text generator 41 recognizes the number of times the selected template has been used by referring to the sentence history. Then, it is determined whether or not the number of times of use is equal to or greater than a predetermined threshold. If it is determined in step S13 that the number of times the selected template has been used is not equal to or greater than the predetermined threshold, step S14 is skipped and step S15 is performed.
The text generation unit 41 generates a message as described above using the selected template,
To return.

If it is determined in step S13 that the number of times the selected template has been used is equal to or greater than the predetermined threshold value, the flow advances to step S14 to output a synthesized sound via the voice synthesizing unit 7 and the speaker 8, for example. By
The user is prompted for a new template,
For example, the user operates a keyboard or the like (not shown) to wait for a new template to be input, and then proceeds to step S15.

In this case, in step S15, a message is generated in the text generating section 41 using the new template input as described above, and the process returns.

[0098] For the template, for example,
Various templates are stored in a predetermined server,
If necessary, the template can be downloaded from the server and used.

As described above, the text generation unit 41 of the sentence generation unit 6 generates a message using a template that has been used less frequently and has not been used recently.
It is possible to prevent a message having a similar phrase that is duplicated with a message transmitted in the past from being transmitted to the transmission partner.

Next, to prevent transmission of a message having a similar phrase as described above, the sentence generation unit 6
In step S4 in FIG. 12, it is also possible to perform the processing according to the flowchart in FIG.

That is, also in this case, steps S21 to S21
At 23, the same processing is performed as in steps S11 to S13 in FIG.

If it is determined in step S23 that the number of times the selected template has been used is not greater than or equal to the predetermined threshold value, steps S24 and S25 are skipped and the process proceeds to step S26, where the text generating unit 41 Used to generate a message and return as described above.

If it is determined in step S23 that the number of times the selected template has been used is equal to or greater than the predetermined threshold, the flow advances to step S24 to replace the predetermined word arranged in the selected template with a new word.

That is, as described with reference to FIG.
In the template, words that can be replaced on the device side (words inserted at the Topic position in the embodiment of FIG. 10) are provided. In step S24, the replaceable words in the selected template are replaced with other words. Is replaced by the word

Then, the process proceeds to step S25, where the text generation unit 41 resets the number of times of use of the selected template in the sentence history to 0, and proceeds to step S26. In step S26, the text generating unit 41 generates a message as described above using the selected template in which some words have been replaced, and returns.

As described above, by replacing a replaceable word of a template used for generating a message with another new word, it is possible to prevent transmission of a message having a similar phrase.

Next, the above-described series of processing can be performed by hardware or can be performed by software. When a series of processing is performed by software, a program constituting the software is
Installed on a general-purpose computer.

FIG. 16 shows an example of the configuration of an embodiment of a computer in which a program for executing the above-described series of processing is installed.

The program is stored in a hard disk 105 or a ROM 1 as a recording medium built in the computer.
03 can be recorded in advance.

Alternatively, the program may be a floppy disk, CD-ROM (Compact Disc Read Only Memory), M
O (Magneto optical) disc, DVD (Digital Versatile)
Disc), a magnetic disk, a semiconductor memory, or another such removable storage medium 111, which can be temporarily or permanently stored (recorded). Such a removable recording medium 111 can be provided as so-called package software.

The program can be installed on the computer from the removable recording medium 111 as described above, can be wirelessly transferred from a download site to the computer via a digital satellite broadcasting artificial satellite, or can be connected to a LAN (Local Area). Network), the Internet, and the like, and can be transferred to a computer by wire. In the computer, the transferred program can be received by the communication unit 108 and installed on the built-in hard disk 105.

The computer has a CPU (Central Processing).
Unit) 102. The CPU 102 has a bus 1
01, an input / output interface 110 is connected. The CPU 102 receives a command via the input / output interface 110 by operating the input unit 107 including a keyboard, a mouse, and the like. Then, according to it, ROM (Read Only Memory)
The program stored in 103 is executed. Alternatively, the CPU 102 transmits the program stored in the hard disk 105, a satellite, or a network, receives the program by the communication unit 108, and
The program installed in the hard disk 105 is read from the removable recording medium 111 installed in the drive 109 and loaded into the RAM (Random Access Memory) 104 and executed. Accordingly, the CPU 102 performs the processing according to the above-described flowchart or the processing performed by the configuration of the above-described block diagram. And
The CPU 102 transmits the processing result to an LCD (Liquid
Output unit 1 consisting of CryStal Display) and speakers
06, or transmitted from the communication unit 108, and further recorded on the hard disk 105.

Here, in the present specification, processing steps for writing a program for causing a computer to perform various types of processing do not necessarily have to be processed in chronological order in the order described in the flowchart. Alternatively, it also includes processing executed individually (for example, parallel processing or processing by an object).

The program may be processed by one computer, or may be processed in a distributed manner by a plurality of computers. Further, the program may be transferred to a remote computer and executed.

In the present embodiment, the message is
Although the synthesized sound is output, the message can be displayed on a display, for example.

Further, in the present embodiment, the speech of the user is recognized by voice and a message is generated based on the voice recognition result. However, the message is generated, for example, by operating the keyboard or the like by the user. It can also be performed based on a text input by the user or a recognition result obtained by recognizing a handwritten input by the user.

In the present embodiment, the case where the present invention is applied to a robot as a real object has been described. However, the present invention is applied to a character virtually realized on a computer, for example. It is also possible.

[0118]

According to the information processing apparatus, the information processing method, and the recording medium of the present invention, an input from a user is analyzed, and at least a partner to which a message is to be transmitted and contents to be transmitted are recognized. Then, a message having contents to be transmitted is generated by an expression according to the intimacy with the partner to which the message is transmitted. Therefore, it is possible to easily obtain a message with an appropriate expression for the other party.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an example of an external configuration of a robot to which the present invention has been applied.

FIG. 2 is a diagram for explaining processing of the robot in FIG. 1;

FIG. 3 is a block diagram showing an example of an electrical configuration of the robot shown in FIG. 1;

FIG. 4 is a block diagram illustrating a configuration example of a voice recognition unit 1;

FIG. 5 is a block diagram illustrating a configuration example of a language processing unit 2.

FIG. 6 is a block diagram illustrating a configuration example of an intention understanding unit 3;

FIG. 7 is a diagram showing an example of person data.

FIG. 8 is a block diagram illustrating a configuration example of a sentence generation unit 6;

FIG. 9 is a diagram illustrating an example of a sentence history.

FIG. 10 is a diagram illustrating an example of a template.

FIG. 11 is a block diagram illustrating a configuration example of a speech synthesis unit 7;

FIG. 12 is a flowchart illustrating a process of the robot in FIG. 3;

FIG. 13 is a diagram showing a result of processing by the robot in FIG. 3;

FIG. 14 is a flowchart illustrating a process of a sentence generation unit 6;

FIG. 15 is a flowchart illustrating another process of the sentence generation unit 6;

FIG. 16 is a block diagram illustrating a configuration example of a computer according to an embodiment of the present invention.

[Explanation of symbols]

1 speech recognition unit, 2 language processing unit, 3 intention understanding unit, 4 control unit, 5 drive unit, 6 sentence generation unit, 7
Voice synthesis unit, 8 speakers, 9 communication unit, 11
Microphone, 12 A / D conversion unit, 13 feature extraction unit, 14 matching unit, 15 acoustic model database, 16 dictionary database, 17 grammar database, 21 text analysis unit, 22 syntax / semantic analysis unit, 23 dictionary database, 24 for analysis Grammar database, 25 Dictionary database, 26
Analysis grammar database, 27 command dictionary database, 31 keyword extraction unit, 32 intention recognition unit, 33 dictionary database, 34 person database, 41 text generation unit, 42 sentence history storage unit, 43 template database, 44 generation grammar database, 45 dictionaries Database, 51
Text analysis unit, 52 rule synthesis unit, 53 D / A
Converter, 54 dictionary database, 55 grammar database for analysis, 56 phoneme segment database, 10
1 bus, 102 CPU, 103 ROM, 104 RA
M, 105 hard disk, 106 output unit,
107 input unit, 108 communication unit, 109 drive, 110 input / output interface, 111 removable recording medium

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (Reference) G10L 15/28 G10L 3/00 551H 9A001 // A63H 11/00 571H (72) Inventor Seiichi Aoyagi Kita Shinagawa, Shinagawa-ku, Tokyo 6-7-35 Inside Sony Corporation (72) Inventor Sachi Tanaka 6-35 7-35 Kita Shinagawa, Shinagawa-ku, Tokyo (72) Inventor Toshio Oe 6-Chome Kitashinagawa, Shinagawa-ku, Tokyo No. 7-35 Sony Corporation F-term (reference) 2C150 BA06 BA11 CA01 CA02 DA05 DA24 DA26 DA27 DA28 DF03 DF04 DF06 DF08 DF33 ED42 ED52 EF13 EF16 EF23 EF29 EF33 EF36 3F060 AA00 BA10 CA14 5B091 AA15 CC21 CB02 5D015 AA05 KK02 KK04 LL12 5D045 AB11 9A001 BB06 CC03 HH13 HH16 HH17 HH18 HH19 JJ12

Claims (11)

[Claims] 1. An information processing apparatus for generating a message for a predetermined partner, comprising: an analyzing unit that analyzes input from a user and recognizes at least a partner to which the message is to be transmitted and contents to be transmitted; An information processing apparatus comprising: a message generation unit that generates a message having the content to be transmitted in an expression according to the degree of intimacy with a partner to which the message is transmitted. 2. The information processing apparatus according to claim 1, further comprising a transmission unit that transmits the message generated by the message generation unit to a partner that transmits the message. 3. The information processing apparatus according to claim 2, wherein the transmitting unit outputs the message as a synthesized sound. 4. The communication device according to claim 2, wherein the transmitting unit transmits the message via a network.
An information processing apparatus according to claim 1. 5. The information processing device according to claim 1, wherein the information processing device is a robot. 6. The information processing apparatus according to claim 1, wherein the message generation unit generates a message that does not overlap with a message generated in the past. 7. A storage unit for storing a history of messages generated by the message generation unit, wherein the message generation unit does not overlap with a message generated in the past by referring to storage contents of the storage unit. 7. The method according to claim 6, wherein a message is generated.
An information processing apparatus according to claim 1. 8. The message generating means replaces one or more words of a message generated in the past,
7. The information processing apparatus according to claim 6, wherein a message that does not overlap with the message is generated. 9. The information processing apparatus according to claim 1, further comprising voice recognition means for recognizing a user's voice, wherein said analysis means analyzes a voice recognition result by said voice recognition means. 10. An information processing method for generating a message for a predetermined partner, comprising: analyzing an input from a user and at least recognizing a partner to which the message is to be transmitted and contents to be transmitted; A message generating step of generating a message of the content to be transmitted in an expression according to the degree of intimacy with the partner to which the message is transmitted. 11. A recording medium on which a program for causing a computer to perform information processing for generating a message for a predetermined partner is analyzed, wherein an input from a user is analyzed, and a partner who transmits the message and a partner who wants to transmit the message. A program comprising an analysis step of at least recognizing the content, and a message generation step of generating a message of the content to be transmitted by an expression according to the intimacy with a partner to which the message is transmitted is recorded. Characteristic recording medium.