JP2020067584A - Communication device and control program for communication device - Google Patents

Abstract

To provide a communication device, etc., capable of talking with a user with the tone and content that match a set character.SOLUTION: A communication device comprises: an input unit for inputting an input speech that is the speech of a user; a response generation module for generating a response speech sentence to the input speech; a determination filter for determining whether or not the content of the response speech sentence generated by the response generation module matches the temperament of a preset character; a conversion filter for referring to a conversion lookup table and replacing with another a specific word, among the words included in the response speech sentence generated by the response generation module, that pertains to character property addition that causes the impression given to the user by a speech; and an output unit for outputting a response speech sentence in which the specific word is replaced by the conversion filter, the output unit outputting a response speech sentence when the content of the response speech sentence is determined to match by the determination filter.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a communication device and a control program for the communication device.

  BACKGROUND ART There is known a technique of analyzing a user's utterance to grasp the meaning and content, generating a response utterance according to the type, and presenting the response utterance to a user by voice or text (for example, refer to Patent Document 1).

JP, 2010-14028, A

  Characters such as animals are given to the communication device so that the user can easily become familiar with them, but even if the variation of characters is increased, if any character speaks in a uniform tone, it causes a sense of discomfort. I will let you. In addition, even when the set character's temperament does not match the utterance content, a sense of discomfort is generated.

  The present invention has been made to solve such a problem, and provides a communication device or the like capable of interacting with a user with a tone and content suitable for a set character.

  The communication device according to the first aspect of the present invention includes an input unit that inputs an input utterance that is an utterance of a user, a response generation module that generates a response utterance sentence for the input utterance, and a response utterance sentence that is generated by the response generation module. A judgment filter that judges whether or not the content matches the preset character's temperament, and character characteristics that change the impression given to the user by the utterance among the words included in the response utterance sentence generated by the response generation module. A conversion filter that replaces a specific word related to assignment by referring to a conversion lookup table and an output unit that outputs a response utterance sentence in which the specific word is replaced by the conversion filter are provided, and the output unit determines that the determination filters match. When it does, a response utterance sentence is output.

  With such a configuration, the response utterance sentence generated by the response generation module that can be used universally can be individually corrected according to the character setting regardless of the character setting, and thus is suitable for the character setting. It is possible to realize speech in tone. In addition, it becomes possible to select a response utterance sentence that matches the character setting from the response utterance sentences that are generated by the response generation module that can be used for general purposes.

  In the communication device, the conversion filter may include a ending conversion filter that replaces the ending word of the response utterance sentence generated by the response generating module as a specific word. The tone that impresses the character's character often appears in the word at the end as one of them. By thus replacing the specific word at the end of the word, it is possible to efficiently correct the tone of the response utterance sentence.

  Further, in the above communication device, the conversion filter replaces at least a part of at least one of the noun, adjective, adjective verb, and adverb, among the words included in the response utterance generated by the response generation module, as a specific word. , It is better to include a word conversion filter. The tone that impresses the character's character may appear as a unique expression of a noun, an adjective, an adjective, or an adverb. By replacing a specific word of such a part of speech, it is possible to efficiently correct the tone of the response utterance sentence.

  In the above communication device, when the output unit has a voice output unit that outputs the response utterance sentence as a voice and a character output unit that outputs the response utterance sentence as a character, the conversion filter outputs the response utterance sentence. It is preferable to refer to the single conversion lookup table to replace the specific word regardless of whether the voice output unit outputs the voice or the character output unit outputs the character.

  Further, in the above communication device, the conversion filter may replace the specific word of the response utterance sentence generated by the response generation module when it is determined that the determination filters match. With such a processing procedure, it is possible to eliminate waste that is not output as a response utterance sentence after performing the replacement processing of the specific word. In addition, the judgment filter refers to the attribute lookup table in which the attributes corresponding to the disposition for each word that can be used when the response generation module generates the response utterance sentence are referred to, and the content of the response utterance sentence is referred to. However, it may be configured to determine whether or not it matches the disposition. With this configuration, the determination can be performed by a relatively simple process, which contributes to speeding up the process.

  The control program of the communication device according to the second aspect of the present invention includes an input step of inputting an input utterance which is an utterance of a user, a response generation step of generating a response utterance sentence for the input utterance, and a response generated in the response generation step. Change the impression given to the user by the utterance, among the words included in the response utterance generated in the response generation step, and a determination step that determines whether or not the content of the utterance matches the preset character's temperament. A conversion step of replacing a specific word relating to character imparting by referring to a conversion lookup table, and an output step of outputting a response utterance in which the specific word is replaced by the conversion step when it is determined that the determination step matches Causes the computer to execute. According to the communication device in which such a program is executed by the computer, it is possible to realize the utterance of the tone and the content suitable for the character setting, as in the first aspect.

  According to the present invention, it is possible to provide a communication device or the like capable of interacting with a user with a tone and content suitable for a set character.

It is a figure which shows the communication example of a robot and a user. It is a system configuration diagram of a robot. It is an example of a setting file that defines the attributes and temperament of a character. It is an example of the ending conversion look-up table used for ending conversion. It is an example of a phrase conversion lookup table used for phrase conversion. It is a flowchart which shows the process from receiving a user's utterance to responding. It is a figure which shows the example of communication of the robot and user which concern on a 1st modification. It is a figure which shows the smart phone which concerns on a 2nd modification.

  Hereinafter, the present invention will be described through embodiments of the invention, but the invention according to the claims is not limited to the following embodiments. In addition, not all of the configurations described in the embodiments are essential as means for solving the problem.

  FIG. 1 is a diagram illustrating an example of communication between a robot 100 according to the first embodiment and a user. The robot 100 is a communication device that performs a voice conversation with a human being who is a user. The robot 100 is a character device that embodies a character, and may be configured so that the facial expression and the line-of-sight direction of the eye change in accordance with a dialogue.

  The robot 100 imitates an animal in appearance and has a head 110 and a body 120. The microphone 101 is hidden and arranged at any position of the head 110. The microphone 101 has a function as an input unit that inputs a user's uttered voice as an input utterance. A speaker 102 is hidden and arranged at the position of the mouth of the robot 100. The speaker 102 also has a function as an output unit that emits a sound generated by the robot 100. The user feels as if the robot 100 is talking to the voice output from the position of the mouth. In the robot 100 described in the present embodiment, the character setting of "grandpa" is made, and the utterance sentence is adjusted and emitted in a tone that is likely to be present in "grandpa". As shown in the figure, for example, when the user talks to the robot 100, "What happens to the weather today?", The robot 100 responds to the talk and utters "It's sunny Chikumori."

  FIG. 2 is a system configuration diagram of the robot 100. Here, the main elements related to communication with the user will be described, but the robot 100 may include other elements and may include other elements that contribute to communication.

  The robot 100 is provided with a microphone 101, a speaker 102, a control unit 200, a memory 210, a speech database 220, a polarity dictionary 230, and a conversion lookup table 240 (hereinafter, referred to as “conversion LUT 240”) as a main system configuration. There is.

  The control unit 200 is configured by, for example, a CPU, controls the entire robot 100, and executes various calculations. The microphone 101 has a main function of collecting a speech voice of the user. The microphone 101 converts the collected voice of the user's utterance into a voice signal and delivers it to the control unit 200 as an input utterance from the user.

  The memory 210 is a non-volatile storage medium, and for example, a solid state drive is used. The memory 210 stores various parameter values, functions, look-up tables, etc. used for control and calculation, in addition to the control program for controlling the robot 100. In particular, the setting file 211 is stored. The setting file 211 is a file that defines attributes and dispositions relating to character characteristics given to the robot 100. Details will be described later.

  The control unit 200 also operates as a function execution unit responsible for executing each function, and mainly operates as the speech analysis unit 201, the response generation module 202, the judgment filter 203, the conversion filter 204, and the speech control unit 205. The utterance analysis unit 201 analyzes the input utterance received from the microphone 101, converts it into text, and recognizes the utterance content of the user. The utterance analysis unit 201 specifically recognizes the utterance content of the user using a general voice recognition technique. For example, a text analysis is performed on the input utterance, and the utterance content is recognized using a DNN model or a logistic regression model. The speech analysis unit 201 passes the recognized speech content to the response generation module 202.

  The response generation module 202 refers to the utterance database 220 and generates a response utterance sentence suitable for the input utterance. The utterance database 220 is, for example, a database stored in a hard disk drive, and each term organized as a corpus is stored together with reproducible voice data. The response generation module 202 is a general-purpose module that generates a standard utterance response utterance sentence, and the utterance database 220 also stores a term group for generating a standard utterance response utterance sentence. Has been done.

  The determination filter 203 receives the response utterance sentence generated by the response generation module 202, and refers to the polarity dictionary 230 to analyze the content thereof. Then, it is determined whether or not the analysis result matches the set character's temperament. If they match, the response utterance sentence is passed to the conversion filter 204, and if they do not match, the response generation module 202 is requested to generate the response utterance sentence again. When a plurality of words having a dependency relationship are input to the polarity dictionary 230, whether the noun therein has an emotionally positive meaning in the response utterance sentence (hereinafter, referred to as “positive”) or a negative word. It is a dictionary that outputs whether it has meaning (hereinafter referred to as "negative"). The polarity dictionary 230 is stored in, for example, a hard disk drive. Specific processing of the judgment filter 203 will be described later.

  The conversion filter 204 is a conversion filter that replaces, among the words included in the response utterance sentence generated by the response generation module 202, a specific word relating to character property imparting that changes the impression given to the user by utterance by referring to the conversion LUT 240. . The conversion filter 204 includes an ending conversion filter 203a and a word conversion filter 203b. The ending conversion filter 203a replaces the ending word of the response utterance sentence generated by the response generating module 202 with reference to the ending conversion LUT stored in the conversion LUT 240. The wording conversion filter 203b stores, in the conversion LUT 240, some of at least one of the words included in the response utterance generated by the response generation module 202, which is a noun, an adjective, an adjective, or an adverb. The phrase conversion LUT is referenced and replaced. The conversion LUT 240 is stored in, for example, a hard disk drive. The conversion filter 204 delivers the response utterance sentence that has undergone the replacement process to the utterance control unit 205. The specific processing of the conversion filter 204 will be described later.

  The utterance control unit 205 converts the received response utterance sentence into a voice signal and delivers it to the speaker 102. The speaker 102 receives the voice signal converted by the utterance control unit 205 and outputs the output utterance as voice.

  FIG. 3 is an example of a setting file 211 that defines the attributes and temperament of a character. The setting file 211 includes two, as shown in the figure, a definition of a character attribute and an attribute table indicating a disposition attribute of a specific word. The character attribute is, for example, “grandpa”, and it is defined to give the robot 100 the attribute of an elderly man. Various other attributes may be defined. Based on age and sex, "grandma" and "boy" can be defined in addition to "grandpa". Further, if a job type or a position is used as a standard, "Samurai-san", "President", etc. can be defined.

  The attribute table enumerates specific words related to the extroversion, for example, when the robot 100 is given "extroversion", which is a preference for going out or outdoors, and each specific word is labeled as "positive (P)". And “negative (N)” temperament attributes are registered in the table. The specific word to be registered is a word that can be used by the response generation module when generating a response utterance sentence.

  When "extroversion" is given, as shown in the figure, the word is "outgoing" which is related to outdoor activities, "P" is related to "camping", and "PC" is related to indoor activities. , “N” temperament attribute is registered for “model”. In this way, the temperament imparted to the robot 100 is indirectly represented by whether "P" or "N" is registered in the related specific word. In other words, the temperament of the robot 100 is generally defined by what specific words are listed in the setting file and what kind of temperament attribute is registered in each specific word.

  The determination filter 203 first analyzes the dependency relation of the response utterance sentence generated by the response generation module 202. For example, if you receive the sentence "I hate camp because I have a lot of insects" as the response utterance sentence, the dependency is "Camp" → "Many", "I hate" and "Insect" → "Many", " Analyze each combination of "I hate". Then, the words “camp” and “dislike” that are likely to be the subject of this sentence are extracted. When the decision filter 203 inputs these two words into the polarity dictionary 230, it obtains the result that “camp” is “negative”. Next, the judgment filter 203 refers to the setting file 211 and confirms whether “camp” is registered or, if registered, is “P” or “N”. As described above, since “camp” is registered as “P” in the setting file 211, it conflicts with the output result of the polarity dictionary 230. Therefore, the determination filter 203 determines that the content of the response utterance sentence does not match the temperament of the robot 100. When the determination filter 203 determines that the response utterance sentence does not match the disposition of the robot 100, the determination filter 203 requests the response generation module 202 to generate the response utterance sentence again.

  For example, when the sentence “I like going out” is received as the response utterance sentence, each combination of “I” → “like” and “going out” → “like” is analyzed as the modification. Then, the words “out” and “like” that are highly likely to be the subject of this sentence are extracted. When the decision filter 203 inputs these two words into the polarity dictionary 230, it obtains the result that “going out” is “positive”. The determination filter 203 then refers to the setting file 211 and confirms whether “going out” is registered or, if registered, whether it is “P” or “N”. As described above, since “going out” is registered as “P” in the setting file 211, it matches the output result of the polarity dictionary 230. Therefore, the determination filter 203 determines that the content of the response utterance sentence matches the temperament of the robot 100. When the determination filter 203 determines that the response utterance sentence matches the temperament of the robot 100, it passes the response utterance sentence to the conversion filter 204. If none of the words extracted from the response utterance sentence is registered in the attribute table of the setting file 211, the response utterance sentence is passed to the conversion filter 204 as it is.

  FIG. 4 is an example of a word ending conversion LUT used for word ending conversion. As shown in the figure, the word ending conversion LUT includes two of a character attribute indicating a target character and a replacement table listing the replacement of specific endings. As shown in the figure, when the character attribute is designated as "grandpa", the ending conversion filter 203a of the robot 100 in which "grandpa" is defined in the setting file 211 is referred to. That is, the ending conversion LUT is created in advance for each character attribute that can be defined in the setting file 211, and is stored in the conversion LUT 240.

  The replacement table is a table listing combinations of specific endings to be replaced and converted endings after the replacement. The specific ending in the ending conversion LUT is a word used as the ending of a response utterance sentence of standard wording generated by the response generating module 202. The converted ending in the ending conversion LUT is a word that the designated character attribute tends to use as the ending. If the character attribute is "Grandpa", it will be combined as "Jada" (converted ending) for "No but" (specific ending) and "Ja" (converted ending) for "is" (specific ending). Be done.

  The ending conversion filter 203a replaces the specific ending of the response utterance sentence generated by the response generating module 202 with the converted ending by referring to the ending conversion LUT. For example, if the response utterance received by the ending conversion filter 203a is "My friends are elderly, but everyone is fine.", "But" "is" corresponds to the specific ending, so change this to " Replaced with "Ja" and "Ja" and said, "My friends are old people, but everyone is fine." In addition, if "I already have a lot of clothes, I wonder if I don't need it.", Then I will say "I don't need it because I already have a lot of clothes."

  Since the tone that impresses the character of the character often appears in the word at the end of the sentence, it is possible to efficiently correct the tone of the response utterance sentence by replacing the specific word at the end in this way. The user can feel the character of the robot 100 more strongly by the corrected tone.

  FIG. 5 is an example of a wording conversion LUT used for wording conversion. As shown in the figure, the wording conversion LUT includes two, a character attribute indicating a target character and a replacement table listing replacements of specific words. As shown in the figure, when the character attribute is designated as "grandpa", the word conversion filter 203b of the robot 100 in which "grandpa" is defined in the setting file 211 is referred to. That is, the wording conversion LUT is created in advance for each character attribute that can be defined in the setting file 211 and is stored in the conversion LUT 240.

  The replacement table is a table listing combinations of specific words to be replaced and converted words after the replacement. The specific word in the phrase conversion LUT is a part of at least one of a noun, an adjective, an adjective verb, and an adverb among words included in the standard utterance response utterance generated by the response generation module 202. is there. The conversion word in the word conversion LUT is a word corresponding to the conversion word that the designated character attribute tends to use. If the character attribute is "Grandpa", "I" (specific word: noun) for "I" (converted word: noun) and "many" (specific word: adverb) for "tanto" (converted) Words: adverbs), "kawaii" (specific words: adjectives) and "kawaii" (converted words: adjectives) are combined.

  The wording conversion filter 203b replaces the specific word of the response utterance sentence whose word ending conversion filter 203a has replaced the word ending with the word conversion LUT with reference to the wording conversion LUT. For example, if the response utterance received by the phrase conversion filter 203b is "My friends are old people, but everyone is fine.", "I" corresponds to a specific word, and this is the converted word "" "I'm old friends, but everyone's fine." If "I already have a lot of clothes, I don't need it.", "I already have clothes, I don't need it."

  The tone that impresses the character's character may appear as a unique phrase such as a noun, adjective, adjective verb, or adverb as one of them, so by replacing a specific word of such a part of speech, the response utterance is efficiently The tone of can be modified. The user can feel the character of the robot 100 more strongly by the corrected tone.

  Next, a flow of processing of the control unit 200 from receiving a user's utterance to responding will be described. FIG. 6 is a flowchart showing the processing from receiving the user's utterance to responding. The flow in the figure shows the processing from the user uttering one phrase to the robot 100 returning one phrase.

  When the control unit 200 acquires a user utterance via the microphone 101 in step S101, the utterance analysis unit 201 as a functional block analyzes and recognizes the user utterance as an input utterance in step S102. The speech analysis unit 201 passes the recognized speech content to the response generation module 202. In step S103, the response generation module 202 refers to the utterance database 220 to generate a response utterance sentence for the input utterance. The response generation module 202 delivers the generated response utterance sentence to the determination filter 203.

  The determination filter 203 analyzes the content of the response utterance sentence received by referring to the polarity dictionary 230 in step S104. Then, in step S105, it is determined by referring to the attribute table of the setting file 211 whether the analysis result matches the set character's temperament. If it is determined that they do not match, the process returns to step S103 so that the response generation module 202 can generate a response utterance sentence again. If it is determined that they match, the response utterance sentence is passed to the conversion filter 204, and the process proceeds to step S106.

  In step S106, the word ending conversion filter 203a refers to the word ending conversion LUT of the conversion LUT 240, replaces the specific ending of the received response utterance sentence, and passes it to the word turning conversion filter 203b. When the response utterance sentence does not include the specific ending, the received response utterance sentence is passed to the conversion filter 203b as it is.

  In step S107, the phrase conversion filter 203b refers to the phrase conversion LUT of the conversion LUT 240, replaces the specific word of the received response utterance sentence, and delivers it to the utterance control unit 205. When the response utterance sentence does not include the specific word, the received response utterance sentence is delivered to the utterance control unit 205 as it is. The order of the processes of steps S106 and S107 may be reversed.

  In step S108, the utterance control unit 205 converts the received output utterance into a voice signal and causes the speaker 102 to speak. With the above, a series of processing is completed. If there is a utterance from the user again, the same process is repeated.

  In the present embodiment described above, the process of the determination filter 203 is performed and then the process of the conversion filter 204 is performed. By performing the processing in this procedure, it is possible to eliminate waste that is not output as a response utterance sentence after the replacement processing of the specific word. However, when the conversion filter 204 is incorporated in the response generation module 202, the order does not necessarily have to be set.

  Next, a first modified example will be described. FIG. 6 is a diagram showing an example of communication between the robot and the user according to the first modification. In the above-described embodiment, the main body is provided with all the main functional elements so that the robot 100 can independently communicate with the user. However, the robot 100 ′ in the first modified example stores the functional elements related to the calculation in the server. Adopt a configuration that is up to 300.

  For example, when the user talks to the robot 100 ', "What happens to the weather today?", The microphone of the robot 100' captures the voice. The robot 100 'converts the captured voice into a voice signal and transmits the voice signal to the server 300 by wireless communication. The server 300 uses these pieces of information to adjust the voice data of the response voice (in the example of the figure, "Hairy Chikumorija") and transmits it to the robot 100 '. The robot 100 'emits a voice corresponding to the received voice data from the speaker 102.

  Even with the system configuration of the first modified example, communication with the user can be realized as in the above-described embodiment. Further, by consolidating the functions related to the calculation in the server 300, the configuration of the robot 100 ′ can be simplified, and the character characteristics can be fully expressed without providing the robot 100 ′ with a high-performance control chip. Communication can be realized. Further, if the server 300 has a function related to the calculation, it is possible to sequentially respond to the calculation requests from the plurality of robots 100 ', so that the manufacturing cost of the entire system can be reduced.

  Next, a second modification will be described. FIG. 8 is a diagram showing a smartphone 700 according to the second modification. In the above-described embodiment and the first modified example, the robots 100 and 100 ′ embodying the character are the users with whom the user interacts, but in the second modified example, the video character 800 displayed on the smartphone 700 interacts. To be a partner. If the character is embodied as a robot, the user can feel like a pet and have more attachment. However, the character can be more easily expressed on the smartphone 700.

  The system configuration of the smartphone 700 is almost the same as the system configuration of the robot 100 described with reference to FIG. A description of common configurations will be omitted, and different configurations will be described below.

  The smartphone 700 includes a display panel 710, a microphone 711, and a speaker 712. The display panel 710 is, for example, a liquid crystal panel, and displays the video character 800. The microphone 711 is an element that replaces the microphone 101 in the robot 100, and collects the voice uttered by the user. The speaker 712 is an element that replaces the speaker 102 in the robot 100, receives the voice signal converted by the utterance control unit 205, and outputs the output utterance as a voice.

  Further, on the display panel 710, an input text window 721 showing the input utterance, which is the utterance of the user, in characters is displayed. The speech analysis unit 201 converts the input speech into characters to generate an input text window 721 and displays it on the display panel 710. Further, the display panel displays an output text window 722 showing the generated response utterance sentence in characters. The utterance control unit 205 character-converts the response utterance sentence to generate an output text window 722, and displays it on the display panel 710.

  When the text information is displayed in this way, the dialogue can be visually confirmed. Also, by turning off the audio output, the user can enjoy communication without disturbing the surroundings even in a quiet environment. In addition, if the input utterance is given as characters using the character input function of the smartphone 700 instead of as voice, the user can enjoy communication without using voice. In this case, both the input utterance and the output utterance are processed as character information.

  When the speaker 712 functions as a voice output unit and the display panel 710 functions as a character output unit as described above, the conversion filter 204 outputs the response utterance sentence as a sound through the speaker 712 and the display panel. Regardless of the case of outputting as a character by 710, it is good to refer to the single conversion LUT 240 and replace the specific word. With this configuration, it is possible to prevent different response utterances from being generated by referring to different LUTs.

  If the smartphone 700 is made to function as a communication device in this way, the dedicated hardware as a communication device is not required, so that the user can more easily enjoy the interaction with the character. In addition, if it is configured so as to be able to interact with the video character 800 in conjunction with other applications of the smartphone 700, it can be applied to various purposes. The smartphone 700 may be a system in which servers cooperate with each other, as in the first example.

  Moreover, if the smartphone 700 is made to function as a communication device, the appearance of the device does not cause the character to be recognized. Therefore, various video characters 800 can be prepared in the smartphone 700, and the user can select a dialogue partner from among them. In this way, when a plurality of selectable characters are prepared for one communication device, a setting file 211 corresponding to each character may be prepared together.

100, 100 ′ robot, 101 microphone, 102 speaker, 110 head, 120 body, 200 control unit, 201 speech analysis unit, 202 response generation module, 203 judgment filter, 204 conversion filter, 204a inflection conversion filter, 204b wording Conversion filter, 205 speech control unit, 210 memory, 211 setting file, 220 speech DB, 230 polarity dictionary, 240 conversion LUT, 300 server, 700 smartphone, 710 display panel, 711 microphone, 712 speaker, 721 input text window, 722 output Text window, 800 video characters

Claims (7)

An input unit for inputting an input utterance that is a user's utterance,
A response generation module that generates a response utterance sentence for the input utterance;
A determination filter for determining whether or not the content of the response utterance sentence generated by the response generation module matches a preset character disposition,
Among the words included in the response utterance sentence generated by the response generation module, a conversion filter that replaces a specific word relating to character property imparting that changes an impression given to the user by utterance by referring to a conversion lookup table,
An output unit that outputs the response utterance sentence in which the specific word is replaced by the conversion filter,
The said output part is a communication apparatus which outputs the said response utterance sentence, when it determines with the said determination filter matching.   The communication device according to claim 1, wherein the conversion filter includes a ending conversion filter that replaces a word at the end of the response utterance sentence generated by the response generation module as the specific word.   The conversion filter replaces at least a part of at least one of a noun, an adjective, an adjective verb, and an adverb among the words included in the response utterance sentence generated by the response generation module, as the specific word. The communication device according to claim 1, further comprising a filter. The output unit includes a voice output unit that outputs the response utterance sentence as a voice, and a character output unit that outputs the response utterance sentence as a character,
The conversion filter refers to the single conversion lookup table regardless of whether the response utterance is output as a voice by the voice output unit or as a character by the character output unit, and the specific word The communication device according to any one of claims 1 to 3, wherein   The communication device according to claim 1, wherein the conversion filter replaces the specific word of the response utterance sentence generated by the response generation module when the conversion filter determines that the determination filter matches.   The determination filter refers to a setting file in which an attribute corresponding to the temperament for each word that can be used when the response generation module generates the response utterance sentence is referred to, and the response utterance sentence The communication device according to claim 1, wherein the communication device determines whether or not the content matches the temperament. An input step of inputting an input utterance that is a user's utterance,
A response generation step of generating a response utterance sentence for the input utterance,
A determination step of determining whether or not the content of the response utterance generated in the response generation step matches a preset character's temperament;
Of the words included in the response utterance sentence generated in the response generation step, a conversion step of replacing a specific word relating to character property imparting a change in the impression given to the user by utterance by referring to a conversion lookup table,
A control program of a communication device for causing a computer to execute an output step of outputting the response utterance sentence in which the specific word is replaced by the conversion step when it is determined that the determination steps match.

Images