JP6977445B2 - Dialogue device, dialogue method and program - Google Patents

Description

The present invention relates to a technique for generating a response sentence when a robot or the like interacts with a user.

Development of terminals and robots that can interact with users is underway. Further, technological development is underway to enable smooth dialogue processing that the user does not get tired of when these terminals and robots interact with the user. For example, Patent Document 1 describes a dialogue system capable of performing a dialogue process that does not get tired of the user by changing a topic, a word, or the like when the system generates a response sentence.

Japanese Unexamined Patent Publication No. 2001-125900

The dialogue system described in Patent Document 1 provides a user with a timeless dialogue by optimizing the topic of conversation at the time of response by using words and the like frequently used by the user as an evaluation index. ing. However, since the pattern of the sentence that the dialogue system responds to is limited to the one stored in the pattern database in advance, the sentence expression of the response sentence does not change even if the topic can be changed for each user. Therefore, there is a high possibility that the sentence expression will be fixed, and it will be difficult for the user to feel a sense of familiarity. As described above, in the conventional dialogue device, there is room for improvement in the technique for making the user feel a sense of familiarity.

The present invention has been made in view of the above circumstances, and an object of the present invention is to improve a technique for giving a user a feeling of familiarity.

To achieve the above object, interaction device of the first aspect of the present invention, obtains the user representation characteristic is information indicating characteristics of expressions included a User chromatography The utterance to be language analysis the user utterance sentence a language analysis unit which, based on a history of the user representation features before Symbol language analysis unit acquires the user's utterance sentence history and history storage unit for storing in a storage unit, which is pre-Symbol history storage unit stores a user's utterance sentence history comprising a system response sentence generation unit that generates a system response sentence before SL and system response sentence output section for outputting the system response sentence system response sentence generation unit has generated, the Te, the system response sentence generation unit, the history A system response sentence is generated using the expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the user speech sentence history stored in the storage unit, and the expression corresponding to the user expression feature is the user expression. It is characterized by being an expression that contrasts with the characteristics.
Further, the dialogue device of the second aspect according to the present invention includes a language analysis unit that linguistically analyzes a user utterance sentence and acquires a user expression feature that is information indicating the feature of the expression included in the user utterance sentence. A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history, and a system response that generates a system response sentence based on the user utterance sentence history stored by the history storage unit. A sentence generation unit and a system response sentence output unit that outputs a system response sentence generated by the system response sentence generation unit are provided, and the language analysis unit determines an utterance type that is information indicating the type of the user utterance sentence. Further acquired, the history storage unit stores the user utterance sentence, the user expression feature, and the history of the utterance type as the user utterance sentence history in the storage unit, and the system response sentence generation unit is the history storage unit. The appearance frequency of the user expression feature can be derived for each utterance type from the user utterance sentence history stored in the above, and the appearance frequency of the user expression feature in the utterance type acquired by the language analysis unit is equal to or higher than the reference frequency. It is characterized in that a system response statement is generated using an expression corresponding to the user expression feature.
Further, the dialogue device of the third aspect according to the present invention includes a language analysis unit that linguistically analyzes a user utterance sentence and acquires a user expression feature that is information indicating the feature of the expression included in the user utterance sentence. A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history, and a system response that generates a system response sentence based on the user utterance sentence history stored by the history storage unit. The sentence generation unit includes a system response sentence output unit that outputs a system response sentence generated by the system response sentence generation unit, and a user identification unit that identifies a user. The history storage unit includes the user identification unit. The user utterance sentence history is stored in the storage unit for each identified user, and the system response sentence generation unit is the user identified by the user identification unit in the user utterance sentence history stored in the history storage unit. It is characterized in that a system response statement is generated using an expression corresponding to the user expression feature whose appearance frequency of the user expression feature is equal to or higher than the reference frequency.
Further, the dialogue device of the fourth aspect according to the present invention includes a language analysis unit that linguistically analyzes a user utterance sentence and acquires a user expression feature that is information indicating the feature of the expression included in the user utterance sentence. A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history, and a system response that generates a system response sentence based on the user utterance sentence history stored by the history storage unit. A sentence generation unit, a system response sentence output unit that outputs a system response sentence generated by the system response sentence generation unit, a user identification unit that identifies a user, an utterance number counting unit that counts the number of utterances of each user, and an utterance count unit. The history storage unit stores the user utterance sentence history for each user identified by the user identification unit in the storage unit, and the system response sentence generation unit stores the user utterance sentence history stored in the history storage unit. Among the above, the system response statement is generated using the expression corresponding to the user expression feature in which the appearance frequency of the user expression feature of the user with the largest number of utterances counted by the utterance number counting unit is equal to or higher than the reference frequency. It is characterized by.
Further, the dialogue method of the first aspect according to the present invention is a dialogue method executed by the dialogue device, which is information indicating the characteristics of the expression included in the user speech by linguistically analyzing the user speech. A language analysis step for acquiring expression features, a history storage step for storing the history of the user expression features acquired in the language analysis step in a storage unit as a user utterance sentence history, and a user utterance stored in the history storage step. The system response statement generation step including a system response statement generation step for generating a system response statement based on a statement history and a system response statement output step for outputting the system response statement generated in the system response statement generation step. Generates a system response sentence using the expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage step, and the expression corresponding to the user expression feature is , The expression is in contrast to the user expression feature.
Further, the dialogue method of the second aspect according to the present invention is a dialogue method executed by the dialogue device, which is information indicating the characteristics of the expression included in the user utterance sentence by linguistically analyzing the user utterance sentence. A language analysis step for acquiring expression features, a history storage step for storing the history of the user expression features acquired in the language analysis step as a user utterance sentence history, and a user utterance stored in the history storage step. The language analysis step includes a system response statement generation step for generating a system response statement based on a statement history and a system response statement output step for outputting the system response statement generated in the system response statement generation step. The utterance type, which is information indicating the type of the user utterance sentence, is further acquired, and the history storage step stores the user utterance sentence, the user expression feature, and the history of the utterance type in the storage unit as the user utterance sentence history. The system response sentence generation step can derive the appearance frequency of the user expression feature for each utterance type from the user utterance sentence history stored in the history storage step, and the utterance type acquired in the language analysis step. It is characterized in that a system response statement is generated using an expression corresponding to the user expression feature in which the appearance frequency of the user expression feature is equal to or higher than the reference frequency.
Further, the dialogue method of the third aspect according to the present invention is a dialogue method executed by the dialogue device, which is information indicating the characteristics of the expression included in the user utterance sentence by linguistically analyzing the user utterance sentence. A language analysis step for acquiring expression features, a history storage step for storing the history of the user expression features acquired in the language analysis step in a storage unit as a user utterance sentence history, and a user utterance stored in the history storage step. A system response statement generation step that generates a system response statement based on the statement history, a system response statement output step that outputs the system response statement generated in the system response statement generation step, a user identification step that identifies a user, and a user identification step. The history storage step stores the user utterance sentence history for each user identified in the user identification step in the storage unit, and the system response sentence generation step stores the user utterance sentence history stored in the history storage step. Among the above, a system response statement is generated using an expression corresponding to the user expression feature in which the appearance frequency of the user expression feature of the user identified in the user identification step is equal to or higher than the reference frequency.
Further, the dialogue method of the fourth aspect according to the present invention is a dialogue method executed by the dialogue device, which is information indicating the characteristics of the expression included in the user utterance sentence by linguistically analyzing the user utterance sentence. A language analysis step for acquiring expression features, a history storage step for storing the history of the user expression features acquired in the language analysis step as a user utterance sentence history, and a user utterance stored in the history storage step. A system response statement generation step that generates a system response statement based on the statement history, a system response statement output step that outputs the system response statement generated in the system response statement generation step, a user identification step that identifies a user, and a user identification step. The history storage step includes a utterance count counting step for counting the number of utterances of each user, and the history storage step stores the user utterance sentence history for each user identified in the user identification step in the storage unit, and generates the system response sentence. The step is the user expression feature in which the appearance frequency of the user expression feature of the user having the largest number of utterances counted in the utterance count counting step in the user utterance sentence history stored in the history storage step is equal to or higher than the reference frequency. It is characterized by generating a system response statement using the corresponding expression.
Further, the program of the first aspect according to the present invention is a language analysis step of acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence by language analysis of the user utterance sentence on a computer. A system response sentence is generated based on a history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history and a user utterance sentence history stored in the history storage step. A user who executes a system response statement generation step and a system response statement output step that outputs a system response statement generated in the system response statement generation step, and stores the system response statement generation step in the history storage step. A system response sentence is generated using an expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the utterance sentence history, and the expression corresponding to the user expression feature is an expression contrasting with the user expression feature. It is characterized by being.
Further, the program of the second aspect according to the present invention is a language analysis step of linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence on a computer. A history storage step for storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history, and a system response sentence are generated based on the user utterance sentence history stored in the history storage step. The system response sentence generation step and the system response sentence output step that outputs the system response sentence generated in the system response sentence generation step are executed, and the language analysis step is information indicating the type of the user utterance sentence. Further acquiring a certain utterance type, the history storage step stores the user utterance sentence, the user expression feature, and the history of the utterance type in the storage unit as the user utterance sentence history, and the system response sentence generation step The appearance frequency of the user expression feature can be derived for each utterance type from the user utterance sentence history stored in the history storage step, and the appearance frequency of the user expression feature in the utterance type acquired in the language analysis step is used as a reference. It is characterized in that a system response statement is generated using an expression corresponding to the user expression feature having a frequency or higher.
Further, the program of the third aspect according to the present invention is a language analysis step of acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance by linguistically analyzing the user utterance to the computer. A system response sentence is generated based on a history storage step in which the history of the user expression feature acquired in the language analysis step is stored in the storage unit as a user utterance sentence history, and a user utterance sentence history stored in the history storage step. The system response statement generation step, the system response statement output step for outputting the system response statement generated in the system response statement generation step, and the user identification step for identifying the user are executed, and the history storage step is the user. The user utterance sentence history is stored in the storage unit for each user identified in the identification step, and the system response sentence generation step is identified in the user identification step in the user utterance sentence history stored in the history storage step. It is characterized in that a system response statement is generated using an expression corresponding to the user expression feature in which the appearance frequency of the user expression feature of the user is equal to or higher than the reference frequency.
Further, the program of the fourth aspect according to the present invention is a language analysis step of linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence on a computer. A history storage step for storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history, and a system response sentence are generated based on the user utterance sentence history stored in the history storage step. A system response statement generation step, a system response statement output step that outputs the system response statement generated in the system response statement generation step, a user identification step that identifies a user, and an utterance count counting step that counts the number of utterances of each user. , The history storage step stores the user utterance sentence history for each user identified in the user identification step in the storage unit, and the system response sentence generation step stores the user utterance sentence stored in the history storage step. In the sentence history, the system response sentence is generated using the expression corresponding to the user expression feature in which the appearance frequency of the user expression feature of the user with the largest number of utterances counted in the utterance number counting step is equal to or higher than the reference frequency. , Characterized by that.

According to the present invention, it is possible to improve the technique for making the user feel a sense of familiarity.

It is a figure which shows the state of the dialogue between the dialogue apparatus and the user which concerns on embodiment of this invention. It is a figure which shows the appearance of the dialogue apparatus which concerns on embodiment. It is a figure which shows the structure of the dialogue apparatus which concerns on embodiment. It is a figure which shows an example of the user utterance sentence history stored in the storage part of the dialogue apparatus which concerns on embodiment. It is a figure which shows the example of the user utterance sentence of each utterance type. It is a figure which shows an example of the result of the parsing of a user utterance sentence by the language analysis part which concerns on embodiment. It is a figure which shows an example of the result which the language analysis part which concerns on embodiment divides a predicate part into an independent storyteller part and an adjunct word part. It is a flowchart which shows the processing procedure of the language analysis processing which concerns on embodiment. It is a figure which shows an example of the positive-negative polar word dictionary which concerns on embodiment. It is a figure which shows an example of the utterance type determination rule which concerns on embodiment. It is a flowchart which shows the processing procedure of the dialogue control processing which concerns on embodiment.

Hereinafter, embodiments of the present invention will be described with reference to figures and tables. The same or corresponding parts in the figure are designated by the same reference numerals.

(Embodiment)
As shown in FIG. 1, the dialogue device 100 according to the embodiment of the present invention is a robot capable of interacting with the user U by voice. As shown in FIG. 2, the dialogue device 100 includes a head 20 and a body 30. The head 20 of the dialogue device 100 is provided with a microphone 21, a camera 22, a speaker 23, and a sensor group 24.

A plurality of microphones 21 are provided on the left and right sides of the head 20 and at the positions of the ears on the human face to form an array microphone. The microphone 21 functions as a voice acquisition unit that acquires the voice spoken by the user U around the dialogue device 100 as voice data.

The camera 22 is an image pickup device provided at the center of the front surface of the head 20 and at the position of the nose in the human face. The camera 22 functions as an image acquisition unit for acquiring image data (image data) of the front image of the dialogue device 100, and inputs the acquired image data to the control unit 110 described later. The dialogue device 100 stores the image data (or the feature amount of the image data) of the user's face in advance in the storage unit 120 described later. Then, the dialogue device 100 can identify (user identification) who the dialogue partner is by recognizing the image data of the face acquired by the camera 22 during the dialogue.

The speaker 23 is provided on the lower side of the camera 22 at the position of the mouth of a person's face. The speaker 23 functions as an audio output unit that outputs audio.

The sensor group 24 is provided at the position of the eyes on the human face. The sensor group 24 includes an acceleration sensor, an obstacle detection sensor, and the like, and is used for detecting various physical quantities and for attitude control, collision avoidance, safety assurance, and the like of the dialogue device 100.

As shown in FIG. 2, the head 20 and the body 30 of the dialogue device 100 are connected to each other by the neck joint 31 shown by the broken line. The neck joint 31 includes a plurality of motors. By driving these a plurality of motors by the control unit 110 described later, the head 20 of the dialogue device 100 can be rotated in three axes of the vertical direction, the horizontal direction, and the tilting direction. As a result, the dialogue device 100 can perform, for example, a nodding operation.

As shown in FIG. 2, a suspension portion 32 is provided at the lower part of the body 30 of the dialogue device 100. The undercarriage portion 32 includes four wheels and a drive motor. Of the four wheels, two are arranged as front wheels on the front side of the fuselage 30, and the other two are arranged as rear wheels on the rear side of the fuselage 30. As the wheel, for example, an omni wheel, a mecanum wheel, or the like may be used. When the control unit 110, which will be described later, controls the drive motor to rotate the wheels, the dialogue device 100 moves.

Next, the functional configuration of the dialogue device 100 will be described with reference to FIG. As shown in FIG. 3, in addition to the above configuration, the dialogue device 100 includes an operation button 33, a control unit 110, and a storage unit 120.

Although not shown, the operation button 33 is provided at a position on the back of the body 30. The operation buttons 33 are various buttons for operating the dialogue device 100. The operation button 33 includes a power button, a volume control button for the speaker 23, and the like.

The control unit 110 is composed of a CPU (Central Processing Unit) or the like. By executing the program stored in the storage unit 120, the control unit 110 includes a voice recognition unit 111, a language analysis unit 112, a history storage unit 113, a system response sentence generation unit 114, a system response sentence output unit 115, which will be described later. It functions as a user identification unit 116. Further, the control unit 110 has a clock function and a timer function, and can acquire the current time (current date and time) and the elapsed time.

The storage unit 120 is composed of a ROM (Read Only Memory), a RAM (Random Access Memory), and the like, and stores programs executed by the CPU of the control unit 110, various data, and the like. Further, in the storage unit 120, the history of information related to the user utterance sentence is stored as the user utterance sentence history 121 by the history storage unit 113.

As shown in FIG. 4, the user utterance sentence history 121 includes information of "user", "user utterance sentence", "utterance type", "sentence end expression", "presence / absence of particles", and "filler". This indicates that the user indicated by the "user" item has spoken the user utterance sentence indicated by the "user utterance sentence" item. Further, the item of "utterance type" is the utterance type of the user utterance sentence, the item of "sentence end expression" is the sentence end expression of the user utterance sentence, and the item of "presence or absence of particle" is the particle of the user utterance sentence. The presence or absence of "filler" indicates that the filler in the user's utterance is stored in the "filler" item. The utterance type, sentence end expression, presence / absence of particles, and filler will be described later.

Next, each function realized by the control unit 110 will be described. As described above, the control unit 110 executes the program stored in the storage unit 120 to execute the voice recognition unit 111, the language analysis unit 112, the history storage unit 113, the system response sentence generation unit 114, and the system response sentence output. It functions as a unit 115 and a user identification unit 116. Further, the control unit 110 supports a multi-thread function, and can execute a plurality of threads (different processing flows) in parallel.

The voice recognition unit 111 voice-recognizes the voice data acquired by the voice acquisition unit (mic 21) and acquires the user's utterance sentence as text data.

The language analysis unit 112 analyzes the user spoken sentence (text data) acquired by the voice recognition unit 111 by a language analysis process described later. Specifically, the language analysis unit 112 performs morphological analysis and syntactic analysis of the user utterance sentence, and acquires part-speech information, dependency information, and the like of each word in the user utterance sentence. Then, based on these, the language analysis unit 112 acquires information indicating the characteristics of the user's expression (user expression characteristics), and determines the utterance type of the user's utterance sentence. As shown in FIG. 5, the utterance type is information (speech type information) representing the type of user utterance sentence (claim, request, question, request, command, proposal, etc.). Further, as user expression features, there are sentence end expressions, presence / absence of particles, fillers, etc., which will be described later.

The history storage unit 113 stores the user utterance sentence acquired by the voice recognition unit 111 and the user expression feature and utterance type acquired or determined by the language analysis unit 112 as the user utterance sentence history 121 for each user. Store in. The end-of-sentence expressions, the presence or absence of particles, and the fillers of the user expression features will be described below.

The sentence end expression is an expression at the end of the sentence spoken by the user. This method of extracting the end-of-sentence expression will be described using the user-spoken sentence "I want you to tell me a recommended movie" of user A shown in FIG. When the language analysis unit 112 parses this user-spoken sentence, as shown in FIG. 6, the predicate unit (I want you to teach me) can be extracted together with the structure of the sentence. Then, as shown in FIG. 7, the language analysis unit 112 divides the predicate part into an independent storyteller and an adjunct word part, so that the adjunct word part can be extracted as a sentence ending expression of the original user utterance sentence. ..

The presence / absence of a particle here refers to the presence / absence of a particle at the end of the dependency source related to the predicate part of the user utterance sentence. Explaining in FIG. 6, the dependency source related to the predicate part (I want you to teach me) is "movie", and the case particle "o" exists at the end of this, so in this example, "presence or absence of particle". "Is". As described above, the "presence / absence of particles" can be obtained by the language analysis unit 112 analyzing the user utterance sentence and the presence / absence of the particle at the end of the dependency source related to the predicate unit of the user utterance sentence.

The filler is an interjection such as "um" or "well" contained in the user's utterance. Here, particles such as "... ne" and "... come" will be included in the filler. In the example of FIG. 6, since there is no filler, “filler” is “absent” in this example. As another example, in the case of the user utterance sentence of user C shown in FIG. 4, "Ah, recommended movie, tell me", the opening "Ah" is an interjection, so it is extracted as a filler. As described above, the "filler" can be extracted as an interjection or an interjection particle included in the user utterance sentence by the language analysis unit 112 analyzing the user utterance sentence and obtaining the part of speech information.

The system response sentence generation unit 114 generates a system response sentence of a standard expression for the user utterance sentence based on the language analysis result of the user utterance sentence and the user utterance sentence. Then, the system response sentence generation unit 114 converts the system response sentence based on the user expression feature acquired from the user utterance sentence history 121 stored in the history storage unit 113. The system response statement generation unit 114 may generate a system response statement based on the user expression feature from the beginning without generating a system response statement of a standard expression.

A method of converting the system response sentence based on the user expression characteristics of the user who is currently interacting with the system response sentence generation unit 114 will be described. In this process, the system response sentence generation unit 114 uses the sentence ending expression with high frequency of appearance and the particle with high frequency of appearance (the particle at the end of the dependency source related to the predicate part) in the past user utterance sentences. The system response sentence is converted based on the usage pattern (particle usage pattern) and the filler with high application frequency. Here, the particle usage patterns used when determining the frequency of appearance in this embodiment are a pattern using particles (a particle usage pattern with particles) and a pattern without particles (no particles) in order to simplify the process. There are two types of particle usage patterns).

First, the system response sentence generation unit 114 targets the data of the user who is currently interacting with the data included in the user utterance sentence history 121, and all the sentence ends stored in the item of "sentence end expression". Among the expressions, sentence-end expressions whose appearance frequency is equal to or higher than the sentence-end expression reference frequency (for example, 0.2) are extracted as frequent sentence-end expressions. The frequency of appearance is calculated as the number of end-of-sentence expressions divided by the total number of data in the user's data. Here, when a plurality of sentence ending expressions are higher than the sentence ending expression standard frequency, the sentence ending expression having the highest frequency of appearance may be used as the frequent sentence ending expression, or randomly selected from the sentence ending expressions having the sentence ending expression standard frequency or higher. One sentence ending expression may be selected and used as a frequent sentence ending expression.

The sentence end expression reference frequency does not have to be a fixed value. For example, when the total number of data in the user utterance sentence history 121 of the user is n, the number of data increases, for example, the sentence end expression reference frequency is set to 2 / n (however, 1 when n = 1). Therefore, the sentence end expression reference frequency may be reduced. Further, the system response sentence generation unit 114 may extract the sentence end expression having the highest appearance frequency as the frequent sentence end expression, instead of the sentence end expression whose appearance frequency is equal to or higher than the sentence end expression reference frequency.

Then, the system response sentence generation unit 114 converts the sentence end expression of the system response sentence into the frequent sentence end expression extracted according to the frequency of appearance as described above.

Next, the system response sentence generation unit 114 targets the data of the user who is currently interacting with the data included in the user utterance sentence history 121, and the appearance frequency is the particle reference frequency (this value is, for example, 0. (Set to a value of 0.5 or more, such as 7th grade) The usage pattern of particles of 5 or more is extracted as a frequent particle usage pattern. The frequency of appearance is calculated as the number of the particle usage patterns in the user's data ÷ the total number of data.

Specifically, the system response sentence generation unit 114 targets the data of the user who is currently interacting among the data included in the user utterance sentence history 121, and is stored in the item of "presence / absence of particles". Find the number of "yes" and the number of "no". Here, assuming that the number of "yes" is represented by m and the number of "no" is represented by n, the frequency of appearance of particles with particles is m / (m + n), and the frequency of appearance of patterns with particles without particles is n. It can be expressed as / (m + n). Here, if n / (m + n) is higher than the particle reference frequency, the particle usage pattern without particles is extracted as the frequent particle usage pattern, and the system response sentence generation unit 114 uses the system response sentence as the particle (predicate unit). Delete the particle) at the end of the relevant dependency source.

If m / (m + n) is higher than the particle reference frequency, the particle usage pattern with particles is extracted as the frequent particle usage pattern, and the system response sentence generation unit 114 does not delete the particles from the system response sentence (system). Leave the particles in the response sentence as they are). If both m / (m + n) and n / (m + n) are equal to or less than the particle reference frequency, it is determined that the particle usage pattern of the user is a usage pattern in which the presence or absence of particles is mixed. The generation unit 114 deletes the particle (the particle at the end of the dependency source related to the predicate unit) included in the system response sentence with a 50% probability. However, this probability does not have to be limited to 50%, and may be deleted with a probability of, for example, n / m. It is possible to make a system response sentence that more reflects the user's frequency of using particles by deleting it with a probability of n / m.

Next, the system response sentence generation unit 114 targets the data of the user who is currently interacting with the data included in the user utterance sentence history 121, and all the data stored in the item of "filler". Among them, a filler having an appearance frequency of the filler reference frequency (for example, 0.1) or more is extracted as a frequent filler. When a plurality of fillers are extracted, the filler having the highest frequency of appearance may be used as the frequent filler. Then, when the frequent filler is extracted, the system response sentence generation unit 114 inserts the frequent filler into the system response sentence.

The filler reference frequency does not have to be a fixed value. For example, when the total number of data in the user utterance history 121 of the user is n, the filler reference frequency may be reduced as the number of data increases, such as setting the filler reference frequency to 1 / n. .. Further, the system response sentence generation unit 114 may extract the filler having the highest appearance frequency as the frequent filler, instead of the filler whose appearance frequency is equal to or higher than the filler reference frequency.

By converting the system response sentence as described above, the system response sentence generation unit 114 system-replies the characteristics of the user's expression based on the three viewpoints (sentence end expression, particle usage pattern, filler) of the user expression characteristic. It can be reflected in the sentence.

The system response sentence output unit 115 synthesizes the system response sentence generated and converted by the system response sentence generation unit 114 by voice, and outputs the voice of the system response sentence from the speaker 23.

The user identification unit 116 recognizes the image data of the user's face acquired by the camera 22 and identifies the current dialogue partner. The user identification unit 116 may identify the dialogue partner by recognizing the characteristics of the voice from the voice data acquired by the microphone 21. Further, the dialogue partner may be identified by using both the recognition result of the image data acquired by the camera 22 and the recognition result of the voice data acquired by the microphone 21.

The functional configuration of the dialogue device 100 has been described above. Next, the language analysis process by the language analysis unit 112 will be described with reference to FIG. This process is performed every time the voice uttered by the user is recognized and the text data is acquired.

First, the language analysis unit 112 performs morphological analysis and syntactic analysis of the user utterance sentence to acquire part-speech information, dependency information, and the like of each word in the user utterance sentence (step S101). Next, the language analysis unit 112 extracts the adjunct part of the predicate unit of the user-spoken sentence that has been parsed as the sentence end expression of the user-spoken sentence (step S102). Then, the language analysis unit 112 acquires the presence / absence of the particle at the end of the dependency source related to the predicate unit of the user utterance sentence (step S103). Further, the language analysis unit 112 extracts the filler from the user's utterance sentence based on the part of speech information (step S104). By the processing from step S101 to step S104, the user expression feature included in the user utterance sentence is acquired.

Next, the language analysis unit 112 adds positive / negative polarity information to the analysis result acquired in step S101 based on the positive / negative polarity word dictionary (step S105). Here, the positive / negative polar word dictionary is a word expressing a positive expression (for example, like (verb), delicious (adjective), beautiful (adjective verb), etc.) and a negative expression, as shown in FIG. It is a word dictionary in which words (for example, dislike (verb), unpleasant (adjective), ugly (adjectival verb), etc.) are registered.

If the analysis result acquired by the language analysis unit 112 in step S101 contains a word representing a positive expression registered in the positive / negative polar word dictionary, "positive" is added to the analysis result as positive / negative polarity information. Will be done. Further, if the analysis result acquired by the language analysis unit 112 in step S101 contains a word representing a negative expression registered in the positive / negative polar word dictionary, the analysis result contains "negative" as positive / negative polarity information. Is given. If the analysis result acquired by the language analysis unit 112 in step S101 does not include a word registered in the positive / negative polar word dictionary, the positive / negative polarity information is not added to the analysis result.

Then, the language analysis unit 112 applies the utterance type determination rule to the analysis result, determines the utterance type (step S106), and ends the process. Here, the utterance type determination rule is a rule for determining the utterance type of the user utterance sentence based on the part of speech, the inflected form, the positive / negative polarity information, etc. at the end of the user utterance sentence, as shown in FIG. If none of the rules apply, the utterance type of the user's utterance sentence is determined to be "other". However, the utterance type determination rule shown in FIG. 10 is only an example. Further, the type of utterance type is not limited to that shown in FIG. For example, "greeting" (such as "hello", utterance of greeting), "agreement" (such as "I agree", utterance indicating that you agree with the utterance of the interlocutor), "thank you" ("thank you", etc.) , "Thank you utterance", "Apology" ("I'm sorry", etc., utterance expressing gratitude), etc. can be added to the type of utterance type.

By the above language analysis processing, the user expression feature included in the user utterance sentence is acquired, and the utterance type of the user utterance sentence is determined. Next, the dialogue control process executed by the control unit 110 of the dialogue device 100 will be described with reference to FIG. This process is started when the dialogue device 100 is activated.

First, the user identification unit 116 recognizes the image data acquired by the camera 22 and identifies the user (step S201). If the image cannot be recognized, such as when the face is not shown in the image data acquired by the camera 22, the user who could be identified immediately before is regarded as the currently interacting user.

Next, the control unit 110 determines whether or not to respond to the user (step S202). This determination can be made from various points of view. For example, if the dialogue device 100 has not yet heard the voice emitted by the user, it is difficult to respond to the user. In this case, it is determined that the dialogue device 100 does not respond. Further, since the normal dialogue responds alternately to the dialogue partner, it may be determined that the user does not respond this time if it is determined to respond last time, and it may be determined that the user does not respond this time if it is determined not to respond last time.

If the control unit 110 determines that there is no response (step S202; No), the voice recognition unit 111 voice-recognizes the voice data acquired from the microphone 21 and acquires the text data of the user's utterance (step S211). Step S211 is also called a voice recognition step.

Then, the language analysis unit 112 linguistically analyzes the text data acquired in step S211 by the above-mentioned language analysis process (FIG. 8) to acquire the user expression feature included in the user utterance sentence and the utterance type of the user utterance sentence. Is determined (step S212). Step S212 is also referred to as a language analysis step.

Then, the history storage unit 113 captures the user name identified in step S201, the user utterance sentence acquired in step S211 and the utterance type determined in step S212 and the acquired user expression characteristics (sentence end expression, presence / absence of particles, filler). It is added to the user utterance sentence history 121 and stored in the storage unit 120 (step S213). Step S213 is also referred to as a history storage step. Then, the process returns to step S201.

On the other hand, if it is determined in step S202 that the control unit 110 responds (step S202; Yes), the system response sentence generation unit 114 generates a system response sentence for the immediately preceding user utterance sentence (step S221). The system response statement generated here is a standard expression system response statement and can be generated by the prior art. The immediately preceding user utterance sentence can be acquired from the user utterance sentence history 121, but the language analysis result of the user utterance sentence is usually required to generate the system response sentence. Therefore, the control unit 110 stores the result of language analysis by the language analysis unit 112 in step S212 with respect to the immediately preceding user utterance sentence in the storage unit 120 so that the system response sentence generation unit 114 can use it in step S221. It is desirable to keep it.

Then, the system response sentence generation unit 114 converts the system response sentence generated in step S221 into a system response sentence using an expression corresponding to the user expression feature included in the user utterance sentence history 121 stored in the storage unit 120. Convert (step S222). Step S222 is also referred to as a system response statement generation step.

For example, suppose the standard system response statement generated in step S221 is "Tell me your favorite food". And if the dialogue partner is user A, the sentence ending expression is "I want you to", there is a particle, and there is no filler, so the system response sentence is "I want you to tell me your favorite food." Is converted to. Also, if the dialogue partner is user C, the sentence ending expression is "tell me", there is no particle, and the filler is "ah", so the system response sentence is "ah, your favorite food, tell me". Is converted to.

Then, the system response sentence output unit 115 voice-synthesizes the system response sentence generated or converted by the system response sentence generation unit 114, and outputs the system response sentence from the speaker 23 (step S223). Step S223 is also referred to as a system response statement output step. Then, the process returns to step S201.

By the dialogue control process described above, the dialogue device 100 can generate a system response statement that reflects the characteristics of the user's expression. Therefore, the system response sentence emitted from the dialogue device 100 becomes a system response statement that imitates the user's habit for each user, and the dialogue device 100 can realize a dialogue that makes the user feel attachment and familiarity. .. In this way, the dialogue device 100 can improve the technique for making the user feel a sense of familiarity.

(Modification 1)
In the above-described embodiment, the system response sentence generation unit 114 has the highest frequency of appearance among all the sentence end expressions stored in the item of "sentence end expression" among the data included in the user utterance sentence history 121. Extract high sentence ending expressions as frequent sentence ending expressions. However, since the sentence end expression changes depending on the utterance type, the appearance frequency considering the utterance type may be used. Such a modification 1 will be described.

In the first modification, the system response sentence generation unit 114 first narrows down only the data of the user who is currently in dialogue among the data included in the user utterance sentence history 121. Next, the system response sentence generation unit 114 determines the utterance type of the system response sentence generated in step S221 of the dialogue control process (FIG. 11), and among the data included in the user utterance sentence history 121, the utterance type. Narrows down only the same data as the utterance type of the system response sentence. Then, the system response sentence generation unit 114 extracts the sentence end expression having the highest frequency of appearance as the frequent sentence end expression among the sentence end expressions stored in the item of the "sentence end expression" of the data narrowed down in this way.

As described above, by extracting the frequent sentence end expressions in consideration of the utterance type, the modification 1 can generate a system response sentence that imitates the user's expression in detail for each utterance type. It is possible to realize a dialogue that makes the user feel more attached and intimate. As described above, the dialogue device according to the first modification can further improve the technique for making the user feel a sense of familiarity.

(Modification 2)
In the above-described embodiment, the system response sentence spoken by the dialogue device 100 is a system response statement that imitates the user's habit by using the user expression feature of the user during the dialogue. However, it is not limited to this. Since some users do not like to imitate their own habits, the system response statement may be converted into an expression opposite to the user expression feature. Such a modification 2 will be described.

In the second modification, the system response sentence generation unit 114 searches for a sentence end expression different from the frequently appearing sentence end expression of the user during the dialogue from the user utterance sentence history 121 of another user, and searches for the sentence end expression of the system response sentence. Convert to the end of sentence expression. If the sentence ending expression of the system response sentence generated in step 221 of the dialogue control process (FIG. 11) is different from the frequently appearing sentence ending expression of the user during the dialogue, the sentence ending expression is not converted (the sentence ending expression of the standard expression). You can leave it as it is).

Then, the system response sentence generation unit 114 uses the particles in a usage pattern opposite to the usage pattern of the user's particles during the dialogue. That is, the system response sentence generation unit 114 targets the data of the user who is currently interacting with the data included in the user utterance sentence history 121, and is stored in the item of "presence / absence of particles". Find the number of and the number of "nothing". Here, as described above, it is assumed that the number of "yes" is represented by m and the number of "no" is represented by n. If n / (m + n) is higher than the particle reference frequency, the particle usage pattern without particles is the user's particle usage pattern. Do not delete the particle at the end of the dependency source related to (leave the particle in the system response sentence as it is).

If m / (m + n) is higher than the particle reference frequency, the particle usage pattern with particles is the user's particle usage pattern. On the contrary, the system response sentence generation unit 114 deletes the particles from the system response sentence. Further, if both m / (m + n) and n / (m + n) are equal to or less than the particle reference frequency, it is determined that the particle usage pattern of the user is a usage pattern in which the presence or absence of particles is mixed. In this case, m and n are compared, and if n is larger (because the user often omits the particle), the system response sentence generation unit 114 uses the particle (predicate unit) included in the system response sentence. Do not delete the particle at the end of the dependency source related to (leave the particle in the system response sentence as it is). Further, if m is larger (because the user often does not omit the particle), the system response sentence generation unit 114 uses the particle included in the system response sentence (the particle at the end of the dependency source related to the predicate unit). ) Is deleted.

Then, the system response sentence generation unit 114 targets all the data stored in the "filler" item among the data included in the user utterance sentence history 121, targeting the data of the user who is currently in dialogue. investigate. Then, if the appearance frequency of the filler is smaller than the filler reference frequency (for example, 0.1), the filler (for example, "Ah", "Eh", etc.) is inserted into the system utterance sentence. The filler to be inserted here may be selected from the fillers included in the user utterance history 121 of another user.

As described above, by intentionally converting the system utterance sentence into an expression different from the user expression feature, a system utterance sentence that reflects the utterance that contrasts with the user's habit (the user's habit is not reflected) is generated. be able to. Therefore, variant 2 makes it possible to generate a system response statement that is not disliked by users who do not like to imitate their own habits, and also feels familiar to users who do not want to imitate their own habits. It is possible to realize a dialogue that makes you feel. Therefore, the dialogue device according to the second modification can improve the technique for making the user feel a sense of familiarity.

The expression used in the system response statement generated by the dialogue device according to the second modification is different from the expression of the user expression feature, but since a different expression is intentionally selected based on the expression of the user expression feature, the expression is selected. It can be said to be a kind of expression corresponding to the user expression feature.

(Modification 3)
In the above-described embodiment, the system response sentence generation unit 114 targets the data of the user who is currently in dialogue among the data included in the user utterance sentence history 121, and the frequent sentence end expression, the presence / absence of particles, and the frequent filler. Is getting. However, it is not necessary to limit the processing to such processing, and for example, the system response statement may be converted based on the user expression characteristics of the user who has the largest number of user utterances. Such a modification 3 will be described.

In the third modification, the control unit 110 also functions as an utterance number counting unit that counts the number of utterances of the user. The utterance number counting unit refers to the item of "user" of the user utterance sentence history 121 stored in the storage unit 120, and counts the number of user utterance sentences of each user. Then, the user with the largest number of user utterances is selected as the representative user.

Then, the system response sentence generation unit 114 targets the data of the user who is currently in dialogue among the data included in the user utterance sentence history 121 when acquiring the frequent sentence end expression, the presence / absence of particles, and the frequent filler. Instead of doing this, the data of the representative user selected by the utterance counting unit is targeted.

As described above, the dialogue device according to the modified example 3 has the highest dialogue frequency by acquiring the frequent sentence ending expression, the presence or absence of particles, and the frequent filler of the representative user and converting the system response sentence by using them. It responds to the user with a system response statement that imitates the user's habit. Then, it becomes possible to motivate the user to have a dialogue with the user who wants to imitate his / her own habit more than others, and it is possible to make each user feel a sense of intimacy. Therefore, the dialogue device according to the third modification can improve the technique for making the user feel a sense of familiarity.

(Other variants)
The present invention is not limited to the above-described embodiment, and various modifications can be made. For example, in the above-described embodiment, the user expression feature is grasped from the three viewpoints of sentence ending expression, presence / absence of particles, and filler, but the present invention is not limited to this. For example, first-person expressions (“I”, “I”, etc.), second-person expressions (“you”, “kun”, etc.), etc. may also be used as user expression features. By doing so, the expression of the user can be more reflected, and the dialogue device can make the user feel more familiar.

Further, when the process of step S221 of the above-mentioned dialogue control process (FIG. 11) is performed, the process of step S222 is also performed, so that the system response sentence of the standard expression is not converted directly. A system response statement that reflects the user expression characteristics of the user may be generated.

Further, in the above-described embodiment, the dialogue device 100 includes a user identification unit 116, user identification is performed by the camera 22 or the microphone 21, and the user utterance sentence history 121 for each user is stored. However, if it is assumed that the dialogue device is used exclusively for a specific user, it is not necessary to include the user identification unit 116. In this case, the camera 22 is also unnecessary, and it is not necessary to provide the item of "user" in the user utterance sentence history 121.

Further, in the above-described embodiment, the user is identified each time in step S201 of the dialogue control process (FIG. 11), but the present invention is not limited to this. Immediately after the dialogue device 100 is started, a user identification process corresponding only to step S201 of the dialogue control process (FIG. 11) is performed to identify the dialogue user, and thereafter, a dialogue control process omitting step S201 is performed. Is also good. Then, in the user identification process, the dialogue device 100 may have the user turn his / her face toward the camera 22 by saying "Please show me your face". Further, after the dialogue device 100 is activated, the user may be identified by voice recognition using the microphone 21 without using the camera 22 by saying "Please tell me your name".

Further, in case the user cannot be identified in any of the user identification processes, the dialogue device 100 defines a user name such as "default user", and the user name of the unidentifiable user is ". It may be treated as "default user" or the like.

Further, in the above-described embodiment, the dialogue device 100 is a movable robot, and various operations can be performed by including the sensor group 24, the neck joint 31, the suspension portion 32, and the like. However, even if the dialogue device cannot perform various operations such as movement, it may be possible to interact with the user. In this case, the sensor group 24, the neck joint 31, and the suspension portion 32 are unnecessary.

Each function of the dialogue device 100 can also be performed by a computer such as a normal PC (Personal Computer). Specifically, in the above embodiment, the program such as the dialogue control process performed by the dialogue device 100 has been described as being stored in the ROM of the storage unit 120 in advance. However, the program is stored and distributed in a computer-readable recording medium such as a flexible disk, a CD-ROM (Compact Disk Read Only Memory), a DVD (Digital Versaille Disc), and an MO (Magnet-Optical Disc), and the program is distributed. You may configure a computer capable of realizing each of the above-mentioned functions by loading and installing the above-mentioned in a computer.

Although the preferred embodiment of the present invention has been described above, the present invention is not limited to the specific embodiment, and the present invention includes the invention described in the claims and the equivalent range thereof. Will be. The inventions described in the original claims of the present application are described below.

(Appendix 1)
A language analysis unit that linguistically analyzes the user's utterance and acquires the user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history in the storage unit, and a history storage unit.
A system response statement generation unit that generates a system response statement based on the user utterance sentence history stored in the history storage unit, and a system response statement generation unit.
A system response statement output unit that outputs a system response statement generated by the system response statement generation unit, and a system response statement output unit.
A dialogue device equipped with.

(Appendix 2)
The system response sentence generation unit generates a system response sentence using an expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage unit.
The dialogue device according to Appendix 1.

(Appendix 3)
The expression corresponding to the user expression feature is the same expression as the user expression feature.
The dialogue device according to Appendix 2.

(Appendix 4)
The expression corresponding to the user expression feature is an expression in contrast to the user expression feature.
The dialogue device according to Appendix 2.

(Appendix 5)
The language analysis unit further acquires the utterance type, which is information indicating the type of the user utterance sentence.
The history storage unit stores the user utterance sentence, the user expression feature, and the history of the utterance type acquired by the language analysis unit as the user utterance sentence history in the storage unit.
The system response sentence generation unit calculates the appearance frequency of the user expression feature for each utterance type from the user utterance sentence history stored in the history storage unit, and the user expression in the utterance type acquired by the language analysis unit. A system response statement is generated using an expression corresponding to the user expression feature whose appearance frequency of the feature is equal to or higher than the reference frequency.
The dialogue device according to any one of Supplementary note 2 to 4.

(Appendix 6)
It also has a user identification unit that identifies the user.
The history storage unit stores the user utterance sentence history for each user identified by the user identification unit in the storage unit.
The system response sentence generation unit corresponds to the user expression feature in which the appearance frequency of the user expression feature of the user identified by the user identification unit is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage unit. Generate a system response statement using the expression
The dialogue device according to any one of Supplementary note 2 to 5.

(Appendix 7)
It also has a utterance count unit that counts the number of utterances of each user.
In the system response sentence generation unit, the appearance frequency of the user expression feature of the user having the largest number of utterances counted by the utterance number counting unit in the user utterance sentence history stored in the history storage unit is equal to or higher than the reference frequency. Generate a system response statement using the expression corresponding to the user expression feature.
The dialogue device according to Appendix 6.

(Appendix 8)
The user expression feature is at least one of a sentence end expression of the user utterance sentence, a particle usage pattern of the user utterance sentence, and a filler included in the user utterance sentence.
The dialogue device according to any one of Supplementary note 1 to 7.

(Appendix 9)
A dialogue method that outputs a system response sentence generated by using a characteristic sentence ending expression, a particle usage pattern, or an expression corresponding to a filler that frequently appears in the user utterance sentence based on the history of the user utterance sentence uttered by the user. ..

(Appendix 10)
A language analysis step for linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence, and a language analysis step.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit, and a history storage step.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and a system response statement generation step.
A system response statement output step that outputs the system response statement generated in the system response statement generation step, and a system response statement output step.
Dialogue methods including.

(Appendix 11)
On the computer
A language analysis step of linguistically analyzing a user's utterance to acquire a user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and
A system response statement output step that outputs the system response statement generated in the system response statement generation step,
A program to execute.

20 ... head, 21 ... microphone, 22 ... camera, 23 ... speaker, 24 ... sensor group, 30 ... torso, 31 ... neck joint, 32 ... undercarriage, 33 ... operation buttons, 100 ... dialogue device, 110 ... control unit , 111 ... Voice recognition unit, 112 ... Language analysis unit, 113 ... History storage unit, 114 ... System response sentence generation unit, 115 ... System response sentence output unit, 116 ... User identification unit, 120 ... Storage unit, 121 ... User utterance Sentence history, U ... user

Claims (15)

A language analysis unit that linguistically analyzes the user's utterance and acquires the user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history in the storage unit, and a history storage unit.
A system response statement generation unit that generates a system response statement based on the user utterance sentence history stored in the history storage unit, and a system response statement generation unit.
A system response statement output unit that outputs a system response statement generated by the system response statement generation unit, and a system response statement output unit.
Equipped with
The system response sentence generation unit generates a system response sentence using an expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage unit.
The expression corresponding to the user expression feature is an expression in contrast to the user expression feature.
A dialogue device characterized by that. A language analysis unit that linguistically analyzes the user's utterance and acquires the user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history in the storage unit, and a history storage unit.
A system response statement generation unit that generates a system response statement based on the user utterance sentence history stored in the history storage unit, and a system response statement generation unit.
A system response statement output unit that outputs a system response statement generated by the system response statement generation unit, and a system response statement output unit.
Equipped with
The language analysis unit further acquires the utterance type, which is information indicating the type of the user utterance sentence.
The history storage unit stores the user utterance sentence, the user expression feature, and the history of the utterance type as the user utterance sentence history in the storage unit.
The system response sentence generation unit is configured to be able to derive the appearance frequency of the user expression feature for each utterance type from the user utterance sentence history stored in the history storage unit, and is configured in the utterance type acquired by the language analysis unit. A system response statement is generated using an expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency.
A dialogue device characterized by that. A language analysis unit that linguistically analyzes the user's utterance and acquires the user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history in the storage unit, and a history storage unit.
A system response statement generation unit that generates a system response statement based on the user utterance sentence history stored in the history storage unit, and a system response statement generation unit.
A system response statement output unit that outputs a system response statement generated by the system response statement generation unit, and a system response statement output unit.
A user identification unit that identifies the user and
Equipped with
The history storage unit stores the user utterance sentence history for each user identified by the user identification unit in the storage unit.
The system response sentence generation unit corresponds to the user expression feature in which the appearance frequency of the user expression feature of the user identified by the user identification unit is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage unit. Generate a system response statement using the expression
A dialogue device characterized by that. A language analysis unit that linguistically analyzes the user's utterance and acquires the user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage unit that stores the history of the user expression feature acquired by the language analysis unit as a user utterance sentence history in the storage unit, and a history storage unit.
A system response statement generation unit that generates a system response statement based on the user utterance sentence history stored in the history storage unit, and a system response statement generation unit.
A system response statement output unit that outputs a system response statement generated by the system response statement generation unit, and a system response statement output unit.
A user identification unit that identifies the user and
An utterance count unit that counts the number of utterances of each user,
Equipped with
The history storage unit stores the user utterance sentence history for each user identified by the user identification unit in the storage unit.
In the system response sentence generation unit, the appearance frequency of the user expression feature of the user having the largest number of utterances counted by the utterance number counting unit in the user utterance sentence history stored in the history storage unit is equal to or higher than the reference frequency. Generate a system response statement using the expression corresponding to the user expression feature.
A dialogue device characterized by that. The expression corresponding to the user expression feature is the same expression as the user expression feature.
The dialogue device according to any one of claims 2 to 4, wherein the dialogue device is characterized. The expression corresponding to the user expression feature is an expression in contrast to the user expression feature.
The dialogue device according to any one of claims 2 to 4, wherein the dialogue device is characterized. The user expression feature is at least one of a sentence end expression of the user utterance sentence, a particle usage pattern of the user utterance sentence, and a filler included in the user utterance sentence.
The dialogue device according to any one of claims 1 to 6, wherein the dialogue device is characterized in that. It is a dialogue method performed by the dialogue device.
A language analysis step for linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence, and a language analysis step.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit, and a history storage step.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and a system response statement generation step.
A system response statement output step that outputs the system response statement generated in the system response statement generation step, and a system response statement output step.
Including
The system response sentence generation step generates a system response sentence using an expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage step.
The expression corresponding to the user expression feature is an expression in contrast to the user expression feature.
A dialogue method characterized by that. It is a dialogue method performed by the dialogue device.
A language analysis step for linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence, and a language analysis step.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit, and a history storage step.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and a system response statement generation step.
A system response statement output step that outputs the system response statement generated in the system response statement generation step, and a system response statement output step.
Including
The language analysis step further acquires the utterance type, which is information indicating the type of the user utterance sentence.
In the history storage step, the user utterance sentence, the user expression feature, and the history of the utterance type are stored in the storage unit as the user utterance sentence history.
In the system response sentence generation step, the appearance frequency of the user expression feature can be derived for each utterance type from the user utterance sentence history stored in the history storage step, and the above-mentioned in the utterance type acquired in the language analysis step. A system response statement is generated using an expression corresponding to the user expression feature whose appearance frequency of the user expression feature is equal to or higher than the reference frequency.
A dialogue method characterized by that. It is a dialogue method performed by the dialogue device.
A language analysis step for linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence, and a language analysis step.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit, and a history storage step.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and a system response statement generation step.
A system response statement output step that outputs the system response statement generated in the system response statement generation step, and a system response statement output step.
A user identification step that identifies a user and
Only including,
In the history storage step, the user utterance sentence history is stored in the storage unit for each user identified in the user identification step.
The system response sentence generation step corresponds to the user expression feature in which the appearance frequency of the user expression feature of the user identified in the user identification step is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage step. Generate a system response statement using the expression
A dialogue method characterized by that. It is a dialogue method performed by the dialogue device.
A language analysis step for linguistically analyzing a user utterance sentence and acquiring a user expression feature which is information indicating the feature of the expression included in the user utterance sentence, and a language analysis step.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit, and a history storage step.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and a system response statement generation step.
A system response statement output step that outputs the system response statement generated in the system response statement generation step, and a system response statement output step.
A user identification step that identifies a user and
An utterance count step that counts the number of utterances of each user,
Including
In the history storage step, the user utterance sentence history is stored in the storage unit for each user identified in the user identification step.
In the system response sentence generation step, in the user utterance sentence history stored in the history storage step, the appearance frequency of the user expression feature of the user having the largest number of utterances counted in the utterance number counting step is equal to or higher than the reference frequency. Generate a system response statement using the expression corresponding to the user expression feature.
A dialogue method characterized by that. On the computer
A language analysis step of linguistically analyzing a user's utterance to acquire a user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit,
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and
A system response statement output step that outputs the system response statement generated in the system response statement generation step,
To execute,
The system response sentence generation step generates a system response sentence using an expression corresponding to the user expression feature whose appearance frequency is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage step.
The expression corresponding to the user expression feature is an expression in contrast to the user expression feature.
A program characterized by that. On the computer
A language analysis step of linguistically analyzing a user's utterance to acquire a user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit.
A system response statement generation step that generates a system response statement based on the user utterance sentence history stored in the history storage step, and
A system response statement output step that outputs the system response statement generated in the system response statement generation step,
To execute,
The language analysis step further acquires the utterance type, which is information indicating the type of the user utterance sentence.
In the history storage step, the user utterance sentence, the user expression feature, and the history of the utterance type are stored in the storage unit as the user utterance sentence history.
In the system response sentence generation step, the appearance frequency of the user expression feature can be derived for each utterance type from the user utterance sentence history stored in the history storage step, and the above-mentioned in the utterance type acquired in the language analysis step. A system response statement is generated using an expression corresponding to the user expression feature whose appearance frequency of the user expression feature is equal to or higher than the reference frequency.
A program characterized by that. On the computer
A language analysis step of linguistically analyzing a user's utterance to acquire a user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit,
A system response statement generation step that generates a system response statement based on the user utterance statement history stored in the history storage step,
A system response statement output step that outputs the system response statement generated in the system response statement generation step, and a system response statement output step.
User identification step to identify the user,
To execute,
In the history storage step, the user utterance sentence history is stored in the storage unit for each user identified in the user identification step.
The system response sentence generation step corresponds to the user expression feature in which the appearance frequency of the user expression feature of the user identified in the user identification step is equal to or higher than the reference frequency in the user utterance sentence history stored in the history storage step. Generate a system response statement using the expression
A program characterized by that. On the computer
A language analysis step of linguistically analyzing a user's utterance to acquire a user's expression feature, which is information indicating the feature of the expression included in the user's utterance.
A history storage step of storing the history of the user expression feature acquired in the language analysis step as a user utterance sentence history in the storage unit,
A system response statement generation step that generates a system response statement based on the user utterance statement history stored in the history storage step,
A system response statement output step that outputs the system response statement generated in the system response statement generation step,
User identification step to identify the user, and
The number of utterances counting step, which counts the number of utterances of each user,
To execute,
In the history storage step, the user utterance sentence history is stored in the storage unit for each user identified in the user identification step.
In the system response sentence generation step, in the user utterance sentence history stored in the history storage step, the appearance frequency of the user expression feature of the user having the largest number of utterances counted in the utterance number counting step is equal to or higher than the reference frequency. Generate a system response statement using the expression corresponding to the user expression feature.
A program characterized by that.

Images