JP4204042B2 - Game machine, game execution method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make a character take action according to the game situation etc. not to repeat a determined speech even if the utterance of a user is not the one previously registered in a database for the conversation control. <P>SOLUTION: In a game system GS comprising a conversation control device 1 for outputting a sentence of reply in response to the utterance of the user and a game device 2 for controlling characters and executing the game, the game device 2 outputs the information describing the game situation to the conversation control device 1, and the conversation control device 1 selects a sentence of reply based on the information describing the game situation informed by the game device 2 and outputs the sentence of reply if topic specifying information suitable for the utterance information can not be found in a plurality of topic specifying information stored in the conversation database 500. <P>COPYRIGHT: (C)2005,JPO&amp;NCIPI

Description

  The present invention relates to a game machine capable of playing a game while having a conversation with a character having a virtual personality, a game execution method, and a program, and more specifically, accepts utterances other than those prepared in advance. It is related with the game machine which can select a reply sentence according to the state of a game, etc., and can implement | achieve natural conversation, the execution method of a game, and a program.

  With the recent development of speech analysis technology and speech recognition technology, conversation systems using a man-machine interface that accepts a user's natural speech as an input and makes a predetermined response to this input are becoming widespread. As an example of such a conversation system, there has been proposed a conversation system that establishes a conversation close to a natural conversation between humans and machines, taking into consideration the history of conversation, transition of topics, etc., rather than just keyword matching ( For example, Patent Document 1).

As business or applied technology using such a conversation system, entertainment (for example, home and business games) using the conversation system can be considered. As an example, it is possible to have a conversation with a character (virtual personality) displayed on a monitor, and to have a game software that advances a game based on the conversation, or to have a conversation with a character while playing a game (the content of the conversation is the outcome of the game) It is considered to incorporate a conversation system as a secondary component in a main game such as game software that does not affect the game.
JP 2002-358304 A

  In the conventional conversation system as described above, the dialogue with the character is determined by the content of the user's utterance. That is, the answer to the user's utterance refers to the contents of a database that stores a plurality of predetermined answers, searches the database for an answer corresponding to the utterance, and returns it to the user. In the technique of simply searching for answers from such a database, the conversation system is a reaction that uniquely utters a predetermined line stored in the database. In such a conventional technique, if the user's utterance is the same, the answer to the utterance is the same, and the same basic speech is repeatedly output as long as the user repeats the same utterance.

  If the answer to the user's utterance is not prepared in the conversation system, for example because the answer corresponding to the user's utterance is not stored in the database, the conversation system performs an unresponsive process, for example, Will be answered, but as long as the user repeats the same utterance, the conversation system repeats the non-response process, for example, "Please tell me again?"

  If such an unresponsive process is repeated, the user feels frustrated and unnatural, which has been a major obstacle to realizing a natural dialogue process between a human and a machine.

  An object of the present invention is to provide a technique for making an answer based on a current game state or the like when a user's utterance is unregistered content.

As means for solving the above problems, the present invention has the following features.
According to a first aspect of the present invention, there is provided a conversation processing means (conversation control apparatus) that outputs an answer sentence in response to an utterance from a user, and a game control means (game apparatus) that controls a character and executes a game. Proposed as a game machine (game system).

  In this game machine, a conversation processing means accepts a user's utterance information (speech recognition unit), a conversation database means (conversation) that stores a plurality of topic identification information and an answer sentence associated with each topic identification information Database), the conversation history determined by the previous conversation and the previous answer sentence, the topic identification information determined by the conversation history is compared with the utterance information, and the topic identification information suitable for the utterance information is selected. Conversation control means (conversation control unit) for outputting an answer sentence associated with the topic identification information, the game control means outputs information indicating the game state to the conversation control means, the conversation control means, If topic specific information suitable for speech information cannot be found from a plurality of topic specific information stored in the conversation database means, based on information indicating the game state Is characterized in that selects an answer sentence.

  According to this game machine, even when an utterance that is not prepared in advance in the conversation database means is accepted, the game machine (whether the user has the initiative (number) of the current game does not repeat the determined line. It is possible to output an answer sentence according to the game state such as whether there is a CPU.

  According to a second aspect of the present invention, there is provided a conversation processing means (conversation control apparatus) for outputting an answer sentence in response to an utterance from a user, a game control means (game apparatus) for controlling a character and executing a game, It is proposed as a game machine (game system) having emotion state information management means for storing emotion state information indicating the emotion of the character and updating the emotion state information according to the utterance information.

  In this game machine, the conversation processing means is a reception means (speech recognition unit) that receives user's utterance information, and a conversation database means (conversation) that stores a plurality of topic identification information and an answer sentence associated with each topic identification information. Database), the conversation history determined by the previous conversation and the previous answer sentence, the topic identification information determined by the conversation history is compared with the utterance information, and the topic identification information suitable for the utterance information is selected. A conversation control unit (conversation control unit) that outputs an answer sentence associated with the topic identification information. The conversation control unit stores a plurality of topic identification information suitable for the utterance information stored in the conversation database unit. When it is not possible to find out from the topic specific information, the answer sentence is selected according to the emotional state information.

  According to this game machine, even when an utterance that is not prepared in advance in the conversation database means is received, it is possible to output an answer sentence according to the character's emotion, instead of repeating the determined line.

  According to a third aspect of the present invention, there is provided a conversation processing means (conversation control apparatus) for outputting an answer sentence in response to an utterance from a user, a game control means (game apparatus) for controlling a character and executing a game, It is proposed as a game machine (game system) having emotion state information management means for storing and updating emotion state information indicating the emotion of the character.

  In this game machine, a conversation processing means accepts a user's utterance information (speech recognition unit), a conversation database means (conversation) that stores a plurality of topic identification information and an answer sentence associated with each topic identification information Database), the conversation history determined by the previous conversation and the previous answer sentence, the topic identification information determined by the conversation history is compared with the utterance information, and the topic identification information suitable for the utterance information is selected. A conversation control unit (conversation control unit) that outputs an answer sentence associated with the topic identification information, and the conversation control unit includes topic identification information suitable for the utterance information stored in the conversation database unit. Stores the unregistered utterance count value, which is the number of times it could not be found from the specific information, and the conversation database means for identifying topic specific information suitable for the utterance information If it is not possible to find from the stored plurality of topic specification information it has been, is characterized by selecting a reply sentence based on the unregistered utterance count.

  According to this game machine, when an utterance that is not prepared in advance in the conversation database means is received, an answer sentence corresponding to the number of times that an unregistered utterance has been performed is output instead of repeating a predetermined line. It becomes possible.

  The invention of the first aspect is also established as a game execution method for executing a game while having a conversation between a user and a character.

  This method includes a step of storing a plurality of topic specifying information and answer sentences associated with each of the topic specifying information, and a conversation history determined by the previous conversation and the previous answer sentence in order to respond to the user's utterance information. The topic identification information suitable for the utterance information is stored in the conversation database means in the steps of selecting the answer sentence by comparing the topic identification information determined by the discourse history with the utterance information and the selecting step described above. A step of selecting an answer sentence based on the game state when the topic identification information cannot be found.

  The invention of the second aspect is also established as a game execution method in which a game is executed while a user and a character have a conversation.

  This method includes a step of storing a plurality of topic specifying information and answer sentences associated with each of the topic specifying information, and a conversation history determined by the previous conversation and the previous answer sentence in order to respond to the user's utterance information. And referring to the topic identification information determined by the discourse history and the utterance information, selecting the answer sentence, storing the emotion state information indicating the emotion of the character, and updating the emotion state information, A step of selecting an answer sentence based on the emotional state information when the topic specifying information suitable for the utterance information cannot be found from the stored topic specifying information in the selecting step described above. Yes.

  In the third aspect of the invention, there is provided a game execution method in which a game is executed while a user and a character are talking, and a plurality of topic specifying information and a response sentence associated with each of the topic specifying information are stored. , Referring to the conversation history determined by the previous conversation and the previous answer sentence in order to respond to the user's utterance information, and comparing the topic identification information determined by the conversation history with the utterance information and selecting the answer sentence; In the selecting step, as a result of collating the topic identification information with the utterance information, the number of times when the topic identification information suitable for the utterance information could not be found from the plurality of stored topic identification information is unregistered In the step of storing the utterance count value and the step of selecting, the topic specifying information suitable for the utterance information is stored from the stored topic specifying information. If you can not out put, it is characterized by a step of selecting a reply sentence based on the unregistered utterance count.

  The invention according to the first to third aspects is also established as a program for causing a computer to function as a game machine and a program for causing a computer to execute a game execution method.

  Here, the “utterance” does not necessarily have to be voice, but may be character string data generated by a character input means such as a keyboard. The “answer sentence” may be made of either a voice or a character string displayed on the screen. In the present invention, the “game” includes not only a game for entertainment but also a game having a purpose other than entertainment such as an educational purpose and a research purpose. The “utterance information” refers to a set of words having a single unit that is uttered by the user, and is, for example, a single sentence (including abbreviated sentences in which some are omitted). “Topic identification information” is information for grasping the theme of conversation between the character played by the user and the game machine.

  According to the present invention, even if the user's utterance is not registered in the database for conversation control in advance, the character is allowed to take action according to the game state or the like instead of repeating the determined dialogue. Therefore, more natural conversation processing can be realized.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present embodiment relates to a game system that can accept a user's voice and make an answer thereto.

[Game system configuration example]
FIG. 1 is a block diagram illustrating a configuration example of a game system. The game system GS includes a conversation control device 1 and a game device 2. The game system GS further includes a voice input unit 3 connected to the conversation control device 1 and the game device 2, a non-voice input unit 4 connected to the game device 2, and an image output unit 5 connected to the game device 2. The voice output means 6 connected to the conversation control apparatus 1 and the game apparatus 2 and the emotion state information management means 7 connected to the conversation control apparatus 1 and the game apparatus 2 are provided.

  The conversation control device 1 has a function of returning an answer according to the user's utterance so that a conversation between the user and the conversation control device 1 is established.

  In addition, the conversation control device 1 changes emotion state information indicating the emotion of the character played by the conversation control device 1 according to the conversation with the user, and outputs an answer sentence corresponding to the emotion of the character.

  The emotion of the character is described by emotion state information. The emotional state information may be any information as long as information indicating emotions can be stored cumulatively. For example, the cumulative value of the emotion flag can be used as the emotional state information.

  As long as the emotion flag is information that can distinguish emotions, any data may be used. For example, the character data “A” is assigned as an emotion flag indicating “normal” emotion, and the emotion flag indicating “furious” emotion is indicated. Character data “B” is assigned, character data “C” is assigned as an emotion flag indicating “anger”, character data “D” is assigned as an emotion flag indicating “joy”, and so on. The conversation control device 1 according to the present embodiment refers to emotion state information stored based on the emotion flag, and controls the emotion of the character (pseudo personality, virtual personality) provided by the conversation control device 1. It becomes.

  The game apparatus 2 has a function of executing a predetermined game process in response to a user input and presenting an execution result to a user (user). Note that any type of game may be handled by the game system GS, but in the present embodiment, a game apparatus 2 that handles a mahjong game that is a competitive game will be described. The game apparatus 2 functions to change the display of the character based on the emotion state information described above. Character display can express not only facial expressions (for example, smiling faces, angry faces, crying faces), but also movements (turning and laughing, waving fists, getting angry, covering faces, crying) and other emotions Everything is included.

  Each of the conversation control device 1 and the game device 2 is an information processing device such as a computer or a workstation, for example, an arithmetic processing unit (CPU), a main memory (RAM), a read only memory (ROM), an input / output device (I). / O), and a device having an external storage device such as a hard disk device. A program for causing an information processing device to function as the conversation control device 1 and / or the game device 2 in the ROM or an external storage device, or a program for causing a computer to execute the conversation control method and / or the game execution method. The conversation control device 1 and / or the game device 2 is realized by placing the program on the main memory and executing it by the CPU. In addition, the program does not necessarily have to be stored in a storage device in the apparatus, and a computer-readable program recording medium such as a magnetic disk, an optical disk, a magneto-optical disk, a CD (Compact Disc), a DVD (Digital Video Disc), etc. Alternatively, it may be provided from an external device (for example, a server of an ASP (Application Service Provider)), and this may be placed on the main memory.

  In FIG. 1, the conversation control device 1 and the game device 2 are displayed as independent devices. However, the conversation control device 1 and the game device 2 do not have to be independent from each other. (CPU), main memory (RAM), read-only memory (ROM), input / output device (I / O), external storage device such as a hard disk device, etc. are used in common, and the conversation control device 1 and the game with the same hardware A configuration for realizing the device 2 may be used.

  The conversation control device 1 and the game device 2 are information processing devices that are physically independent from each other, and are connected via a network to exchange information with each other to constitute a game system GS. It doesn't matter.

  The voice input means 3 has a function of converting a user's voice signal into a predetermined signal such as an electric signal or an optical signal and supplying the signal to the conversation control device 1 and the game device 2, and is a voice input microphone, for example. The voice input means 3 accepts input of the user's utterance to the pseudo personality (character) realized by the conversation control device 1 by a program or the like, and inputs related to the game (for example, pong, reach, or ron) Accept.

  The non-speech input unit 4 has a function of providing the game apparatus 2 with input by the user other than the voice input, and is a keyboard, a joystick, a controller, a pointing device, or the like, for example. Thereby, the user can request the game apparatus 2 to perform input necessary for the game, such as creating a bag, throwing away the bag, pong, reaching, and playing.

  The image output means 5 has a function of displaying a game screen (including character display) generated by the game apparatus 2 as the game progresses to the user, and is, for example, a liquid crystal display apparatus.

  The voice output means 6 outputs the content output as an answer sentence of the character (pseudo personality, virtual personality) by the conversation control device 1 as a voice signal, and also outputs voice / sound (sound effect, BGM, etc.) related to the game. For example, a sound board or a speaker.

  The emotional state information management means 7 receives the emotional flag output from the conversation control device 1, performs a process of reflecting the emotional flag in the emotional state information, returns the current emotional state information to the conversational control device 1, and sends the emotional state. The answer sentence according to the information is output.

  In the configuration example shown in FIG. 1, the emotion state information management unit 7 is displayed as a component independent of the conversation control device 1 and the game device 2, but the emotion state information management unit 7 may be the conversation control device 1 or the game device. 2 may be incorporated, or may be incorporated as a part of these devices.

[Game device]
Next, a configuration example of the game apparatus 2 will be described with reference to FIG.
As described above, the game apparatus 2 includes an external storage device such as an arithmetic processing unit (CPU), a main memory (RAM), a read-only memory (ROM), an input / output device (I / O), and a hard disk device. By executing a predetermined program on the information processing apparatus, the game apparatus 2 and the following components constituting the game apparatus 2 are realized.
FIG. 2 is a block diagram illustrating a configuration example of the game apparatus 2. The game apparatus 2 includes a character string / command conversion unit 201, a game progress control unit 202, an image processing unit 203, and a sound processing unit 204.

  The character string / command conversion unit 201 has a function of converting the user's utterance sent from the conversation control device 1 into character commands, and for example, when the user utters “Pon!” The command conversion unit 201 changes the character string information “Pong” sent from the conversation control device 1 to an execution command of “Pong” and passes it to the game progress control unit 202. It should be noted that, out of the user's speech, those not related to the progress of the game ( "Hello", "name?") For the string / instruction conversion unit 201 is not the any of the output.

  The game progress control unit 202 has a function of causing the game to proceed in accordance with user input and designating data for screen display and data for audio output according to the progress.

  In addition, the game progress control unit 202 receives emotion state information from the emotion state information management means 7, and executes image control of the character in response thereto. For example, if the emotional state information indicates a state in which the character is pleased, the character is controlled so as to show a happy expression or action.

  In addition, the game progress control unit 202 passes the game state information to the conversation control device 1. In the present specification, the “game state information” is information relating to the progress of the game that can be grasped by the game apparatus 2, and as an example, in the game adopting a so-called turn system, it is a user's turn. Or the game device 2 (CPU) side.

The image processing unit 203 reads the screen display data designated by the game progress control unit 202 from a storage unit (not shown) that stores image data necessary for the game in advance, and stores the data in the image output unit 5. provide.
The voice processing unit 204 reads out data for screen display designated by the game progress control unit 202 from a storage unit (not shown) that stores voice data / acoustic data necessary for the game in advance, and outputs this as voice output. Provided to means 6.

[Configuration example of conversation control device]
[overall structure]
FIG. 3 is a functional block diagram illustrating a configuration example of the conversation control device 1 according to the present embodiment.
As described above, the conversation control device 1 includes an external storage device such as an arithmetic processing unit (CPU), a main memory (RAM), a read-only memory (ROM), an input / output device (I / O), and a hard disk device. The following components constituting the conversation control device 1 and the conversation control device 1 are realized by executing a predetermined program by the information processing device.
As shown in FIG. 3, the conversation control device 1 includes a speech recognition unit 200, a conversation control unit 300, a sentence analysis unit 400, a conversation database 500, and a speech recognition dictionary storage unit 600.

[Voice recognition part]
The voice recognition unit 200 specifies a character string corresponding to the utterance content based on the signal corresponding to the utterance provided from the voice input unit 3. Specifically, the voice recognition unit 200 to which the voice signal is input from the voice input unit 3 collates this voice signal with the dictionary and the conversation database 500 stored in the voice recognition dictionary storage unit 600 to obtain the voice signal. The speech recognition result estimated from the above is output. In the configuration example illustrated in FIG. 3, the speech recognition unit 200 requests the conversation control unit 300 to acquire the storage content of the conversation database 500, and the storage content of the conversation database 500 acquired by the conversation control unit 300 in response to the request. However, the voice recognition unit 200 may directly acquire the stored contents of the conversation database 500 and compare it with the voice signal.

[Configuration example of voice recognition unit]
FIG. 4 is a functional block diagram illustrating a configuration example of the voice recognition unit 200. The speech recognition unit 200 includes a feature extraction unit 200A, a buffer memory (BM) 200B, a word matching unit 200C, a buffer memory (BM) 200D, a candidate determination unit 200E, and a word hypothesis narrowing unit 200F. Yes. The word matching unit 200C and the word hypothesis narrowing unit 200F are connected to the speech recognition dictionary storage unit 600, and the candidate determination unit 200E is connected to the conversation control unit 300.

  The speech recognition dictionary storage unit 600 connected to the word matching unit 200C stores a phoneme hidden Markov model (hereinafter, the hidden Markov model is referred to as an HMM). The phoneme HMM is represented including each state, and each state has the following information. (A) state number, (b) acceptable context class, (c) list of preceding and subsequent states, (d) parameters of output probability density distribution, and (e) self-transition probabilities and transitions to subsequent states. It consists of probabilities. Note that the phoneme HMM used in the present embodiment is generated by converting a predetermined speaker mixed HMM because it is necessary to specify which speaker each distribution is derived from. Here, the output probability density function is a mixed Gaussian distribution having a 34-dimensional diagonal covariance matrix. The speech recognition dictionary storage unit 600 connected to the word collation unit 200C stores a word dictionary. The word dictionary stores a symbol string indicating a symbolic reading for each word of the phoneme HMM.

  The voice of the speaker is input to a microphone or the like and converted into an audio signal, and then input to the feature extraction unit 200A. The feature extraction unit 200A performs A / D conversion on the input audio signal, extracts feature parameters, and outputs them. There are various methods for extracting and outputting feature parameters. For example, as an example, LPC analysis is performed, and logarithmic power, 16th-order cepstrum coefficient, Δlogarithmic power, and 16th-order Δcepstrum coefficient are calculated. Examples include a method of extracting 34-dimensional feature parameters. The extracted time series of feature parameters is input to the word matching unit 200C via the buffer memory (BM) 200B.

  The word matching unit 200C uses the one-pass Viterbi decoding method, and based on the feature parameter data input via the buffer memory 200B, the phoneme HMM and the word dictionary stored in the speech recognition dictionary storage unit 600 Is used to detect the word hypothesis, and the likelihood is calculated and output. Here, the word matching unit 200C calculates the likelihood in the word and the likelihood from the start of utterance for each state of each HMM at each time. The likelihood is individually provided for each word identification number, word start time, and difference between preceding words. Further, in order to reduce the amount of calculation processing, the low likelihood grid hypothesis among the total likelihoods calculated based on the phoneme HMM and the word dictionary may be reduced. The word matching unit 200C includes information on the detected word hypothesis and its likelihood along with time information from the utterance start time (specifically, for example, a frame number) and the candidate determination unit 200E and the word hypothesis narrowing down via the buffer memory 200D. Output to the unit 200F.

  The candidate determining unit 200E refers to the conversation control unit 300, compares the detected word hypothesis with the topic specifying information in the predetermined discourse space, and among the detected word hypotheses, the topic specifying information in the predetermined discourse space If there is a match, the matching word hypothesis is output as a recognition result. If there is no match, the word hypothesis narrowing unit 200F Request to narrow down hypotheses.

  An operation example of the candidate determination unit 200E will be described. Now, the word matching unit 200C outputs a plurality of word hypotheses “Kantaku”, “Katak”, “Kantoku” and the likelihood (recognition rate), and the predetermined discourse space relates to “movie”, and the topic specifying information includes “Kantoku (director)” is included, but “Kantaku (reclaimed)” and “Katak (temporary contract)” are not included. Further, the likelihood (recognition rate) of “Kantaku”, “Katak”, and “Kantoku” is highest in “Kantaku”, lowest in “Kantoku”, and “Katak” is in between.

  In the above situation, the candidate determining unit 200E compares the detected word hypothesis with the topic specifying information in the predetermined discourse space, and the word hypothesis “Kantoku” matches the topic specifying information in the predetermined discourse space. The word hypothesis “Kantoku” is output as a recognition result and passed to the conversation control unit. By processing in this way, “Kantoku (Director)” related to the topic “Movie” currently being handled is prioritized over the word hypotheses “Kantaku” and “Katak” with higher likelihood (recognition rate). As a result, it is possible to output a speech recognition result in accordance with the context of the conversation.

  On the other hand, if there is no match, the word hypothesis narrowing unit 200F operates to output the recognition result in response to a request for narrowing down word hypotheses from the candidate determination unit 200E. The word hypothesis narrowing-down unit 200F refers to a statistical language model stored in the speech recognition dictionary storage unit 600 based on a plurality of word hypotheses output from the word matching unit 200C via the buffer memory 200D. For the same word hypothesis of the same word with the same end time but different start time, the highest likelihood of the calculated total likelihood from the utterance start time to the end time of the word for each head phoneme environment of the word After narrowing down word hypotheses so that they are represented by a single word hypothesis having a degree, a word string of a hypothesis having the maximum total likelihood is output as a recognition result among the word strings of all the word hypotheses after narrowing down To do. In the present embodiment, preferably, the first phoneme environment of the word to be processed is three phonemes including the final phoneme of the word hypothesis preceding the word and the first two phonemes of the word hypothesis of the word. Say a line.

An example of word narrowing processing by the word hypothesis narrowing unit 200F will be described with reference to FIG. FIG. 5 is a timing chart showing an example of processing of the word hypothesis narrowing-down unit 200F.
For example, when the i-th word Wi consisting of the phoneme sequence a1, a2,..., An comes after the (i-1) -th word Wi-1, six hypotheses Wa, Assume that Wb, Wc, Wd, We, and Wf exist. Here, it is assumed that the final phoneme of the former three word hypotheses Wa, Wb, and Wc is / x /, and the final phoneme of the latter three word hypotheses Wd, We, and Wf is / y /. Assuming that three hypotheses premised on the word hypotheses Wa, Wb, and Wc and one hypothesis premised on the word hypotheses Wd, We, and Wf remain at the end time te, the former three with the same initial phoneme environment Of the hypotheses, one of the hypotheses with the highest total likelihood is left and the others are deleted.

  Note that the hypothesis premised on the word hypothesis Wd, We, Wf is that the leading phoneme environment is different from the other three hypotheses, that is, the final phoneme of the preceding word hypothesis is y instead of x, so this word hypothesis Wd , We and Wf are not deleted. That is, only one hypothesis is left for each final phoneme of the preceding word hypothesis.

In the above embodiment, the first phoneme environment of the word is defined as three phoneme sequences including the final phoneme of the word hypothesis preceding the word and the first two phonemes of the word hypothesis of the word. However, the present invention is not limited to this, the phoneme string of the preceding word hypothesis including the final phoneme of the preceding word hypothesis, and at least one phoneme of the preceding word hypothesis continuous with the last phoneme, and the word hypothesis of the word A phoneme sequence including a phoneme string including the first phoneme may be used.
In the above embodiment, the feature extraction unit 200A, the word collation unit 200C, the candidate determination unit 200E, and the word hypothesis narrowing unit 200F are configured by a computer such as a digital electronic computer, for example, and include a buffer memory 200B, The 200D and the speech recognition dictionary storage unit 600 are configured by a storage device such as a hard disk memory, for example.

In the above embodiment, speech recognition is performed using the word matching unit 200C and the word hypothesis narrowing unit 200F. However, the present invention is not limited to this, for example, a phoneme matching unit that refers to a phoneme HMM; For example, One
You may comprise with the speech recognition part which performs speech recognition of a word with reference to a statistical language model using a Pass DP algorithm.
In the present embodiment, the voice recognition unit 200 is described as a part of the conversation control device 1, but an independent voice recognition device including the voice recognition unit 200, the voice recognition dictionary storage unit 600, and the conversation database 500. It is also possible.

[Operation example of voice recognition unit]
Next, the operation of the speech recognition unit 200 will be described with reference to FIG. FIG. 6 is a flowchart showing an operation example of the speech recognition unit 200. When receiving a voice signal from the voice input unit 3, the voice recognition unit 200 performs a feature analysis of the input voice and generates a feature parameter (step S601). Next, this feature parameter is compared with the phoneme HMM and language model stored in the speech recognition dictionary storage unit 600 to obtain a predetermined number of word hypotheses and their likelihoods (step S602). Next, the speech recognition unit 200 compares the acquired predetermined number of word hypotheses with the detected word hypothesis and the topic identification information in the predetermined discourse space, and among the detected word hypotheses, the topic in the predetermined discourse space It is determined whether there is a match with the specific information (steps S603 and S604). If there is a match, the speech recognition unit 200 outputs the matching word hypothesis as a recognition result (step S605). On the other hand, when there is no match, the speech recognition unit 200 outputs the word hypothesis having the maximum likelihood as the recognition result according to the acquired likelihood of the word hypothesis (step S606).

[Voice recognition dictionary storage]
Returning to FIG. 3 again, the description of the configuration example of the conversation control device 1 will be continued.
The speech recognition dictionary storage unit 600 stores a character string corresponding to a standard speech signal. The voice recognition unit 200 that has performed the collation identifies a character string corresponding to the word hypothesis corresponding to the voice signal, and outputs the identified character string to the conversation control unit 300 as a character string signal.

[Sentence Analysis Department]
Next, a configuration example of the sentence analysis unit 400 will be described with reference to FIGS. 1 and 7. FIG. 7 is a partial enlarged block diagram of the conversation control device 1, and is a block diagram illustrating a specific configuration example of the conversation control unit 300 and the sentence analysis unit 400. In FIG. 7, only the conversation control unit 300, the sentence analysis unit 400, and the conversation database 500 are shown, and the display of other components is omitted.

  The sentence analysis unit 400 analyzes the character string specified by the speech recognition unit 200. In this embodiment, as shown in FIG. 7, the sentence analysis unit 400 includes a character string identification unit 410, a morpheme extraction unit 420, a morpheme database 430, an input type determination unit 440, and an utterance type database 450. have. The character string specifying unit 410 divides a series of character strings specified by the voice input unit 3 and the voice recognition unit 200 for each sentence. This one-sentence means a delimiter sentence in which character strings are divided as finely as possible without breaking the meaning of the grammar. Specifically, when there is a certain time interval or more in a series of character strings, the character string specifying unit 410 divides the character string at that portion. The character string specifying unit 410 outputs the divided character strings to the morpheme extracting unit 420 and the input type determining unit 440. It should be noted that “character string” described below means a character string for each phrase.

[Morpheme extraction unit]
The morpheme extraction unit 420 sets, as first morpheme information, each morpheme constituting the minimum unit of the character string from the character string of the one phrase according to the character string of the one sentence divided by the character string specifying unit 410. To extract. Here, in this embodiment, the morpheme means the minimum unit of the word structure represented in the character string. Examples of the minimum unit of the word structure include parts of speech such as nouns, adjectives and verbs.

  FIG. 8 is a diagram illustrating a relationship between a character string and a morpheme extracted from the character string. As shown in FIG. 8, each morpheme can be expressed as m1, m2, m3... In the present embodiment. As shown in FIG. 8, the morpheme extraction unit 420 to which the character string is input from the character string specifying unit 410 includes the input character string and a morpheme group stored in advance in the morpheme database 430. Morphemes that belong to the part-of-speech classification are prepared as a morpheme dictionary that describes the morpheme entry word, reading, part-of-speech, utilization form, etc.). The matched morpheme extraction unit 420 extracts each morpheme (m1, m2,...) That matches one of the previously stored morpheme groups from the character string. Examples of the elements (n1, n2, n3...) Excluding each extracted morpheme include auxiliary verbs.

  The morpheme extraction unit 420 outputs each extracted morpheme to the topic identification information search box 320 as first morpheme information. Note that the first morpheme information need not be structured. Here, “structured” means to classify and arrange morphemes contained in a character string based on the part of speech, for example, a character string that is an utterance sentence, such as “subject + object + predicate”. In the same way, it refers to conversion into data obtained by arranging morphemes in a predetermined order. Of course, even if structured first morpheme information is used, this does not interfere with the implementation of the present embodiment.

[Input type determination unit]
The input type determination unit 440 has a function of determining the type of utterance content (speech type) based on the character string specified by the character string specifying unit 410 and outputting information indicating the utterance type according to the determination result. This utterance type is information for specifying the type of utterance content, and in the present embodiment, it means, for example, the “spoken sentence type” shown in FIG. FIG. 9 is a diagram illustrating an example of “spoken sentence type”, a two-letter alphabet representing the type of the spoken sentence, and a spoken sentence corresponding to the type of the spoken sentence.

  In this embodiment, as shown in FIG. 9, “spoken sentence type” is a statement sentence (D; Declaration), a time sentence (T; Time), a location sentence (L; Location), and a repulsive sentence. (N; Negation) and the like, and is used to determine the answer sentence according to the type of the utterance sentence. The sentence composed of each type is composed of an affirmative sentence or a question sentence. The “declaration sentence” means a sentence indicating a user's opinion or idea. In the present embodiment, this statement is a sentence such as “I like Sato” as shown in FIG. “Place sentence” means a sentence with a place concept. “Time sentence” means a sentence with a temporal concept. “Rebound sentence” means a sentence when a statement is denied. An example sentence for “spoken sentence type” is as shown in FIG.

  In the case of determining “spoken sentence type”, in this embodiment, the input type determining unit 440 is a definition expression dictionary for determining that it is a statement sentence, and a repulsive expression for determining that it is a repulsive sentence. A dictionary group describing the correspondence between the expression of the dictionary and the type of utterance is used. FIG. 10 shows a dictionary to be used and types of determinations that are made when the corresponding expression is included in the dictionary. As a specific example, the input type determination unit 440 to which a character string is input from the character string specifying unit 410, based on the input character string, each character string and each dictionary stored in the utterance type database 450, Is matched. The input type determination unit 440 that has performed the collation extracts elements (data indicating the type of determination such as D and N) related to each dictionary from the character string.

  The input type determination unit 440 determines “spoken sentence type” based on the extracted elements. For example, when an element that describes a certain event is included in a character string, the input type determination unit 440 determines the character string that includes the element as a statement. The input type determination unit 440 outputs the determined “spoken sentence type” to the answer acquisition unit 350.

[Conversation database]
Next, a data configuration example of data stored in the conversation database 500 will be described with reference to FIG. FIG. 11 is a conceptual diagram illustrating an example of a data configuration of data stored in the conversation database 500.

  As shown in FIG. 11, the conversation database 500 stores a plurality of pieces of topic specifying information 810 for specifying topics in advance. Each topic specifying information 810 may be associated with other topic specifying information. For example, in the example shown in FIG. 11, when the topic specifying information C is specified, it is associated with the topic specifying information C. The other topic specifying information A, B, D is stored so as to be determined.

  Specifically, in the present embodiment, the topic identification information 810 means “keywords” that are relevant to the input content expected to be input by the user or the answer sentence to the user.

  One or more topic titles 820 are stored in the topic specifying information 810 in association with each other. The topic title 820 is composed of morphemes composed of one character, a plurality of character strings, or a combination thereof. Each topic title 820 stores an answer sentence 830 to the user in association with it. In addition, a plurality of answer types indicating the type of the answer sentence 830 are associated with the answer sentence.

  Each answer sentence 830 is associated with an emotion condition parameter 840. The emotion condition parameter 840 is information indicating a condition of emotion state information. For example, when the emotion state information indicates the cumulative value 10 of the emotion flag “anger”, the emotion condition parameter 840 of a certain response sentence A describes the cumulative value 5 or less of the emotion flag “anger”. When the emotion condition parameter 840 of another answer sentence B describes the cumulative value of the emotion flag of “anger” of 8 or more, the answer sentence A is not selected, and the answer sentence B is sent from the conversation control device 1 to the user. It will be selected as an answer.

  Furthermore, an emotion flag 850 that is data indicating the emotion of the pseudo personality provided by the conversation control device 1 is stored in association with the answer sentence 830. The emotion flag 850 may be any data as long as it can distinguish emotions. For example, the character data “A” is assigned as an emotion flag indicating “normal” emotion and the emotion indicating “anger” emotion is indicated. Character data “B” is assigned as a flag, character data “C” is assigned as an emotion flag 850 indicating “anger”, character data “D” is assigned as an emotion flag indicating “joy”, and so on. To do. The conversation control device 1 according to the present embodiment refers to the emotion flag 850 and controls the emotion of the character (pseudo personality) provided by the conversation control device 1.

Next, the association between certain topic specifying information and other topic specifying information will be described. FIG. 12 is a diagram showing an association between certain topic specifying information 810A and other topic specifying information 810B, 810C _{1 to} 810C ₄ , 810D _{1 to} 810D ₃ . In the following description, “stored in association” means that when information X is read, information Y associated with the information X can be read. For example, information Y in the data of the information X Is stored as information (for example, a pointer indicating the storage destination address of information Y, a physical memory address of the storage destination of information Y, and a logical address). "Remembered".

  In the example shown in FIG. 12, the topic identification information can be stored in association with other topic identification information in association with a higher concept, a lower concept, a synonym, and a synonym (omitted in the example of this figure). In the example shown in this figure, topic specifying information 810B (= “entertainment”) is stored in association with the topic specifying information 810A as topic specifying information of the higher concept for the topic specifying information 810A (= “movie”). The topic specific information (“movie”) is stored in the upper hierarchy.

Further, topic specific information 810C ₁ (= “director”), topic specific information 810C ₂ (= “starring”), topic specific information 810C ₃ (= “distribution company” for the topic specific information 810A (= “movie”) )), Topic identification information 810C ₄ (= “screening time”), topic identification information 810D ₁ (= “Seven Samurai”), topic identification information 810D ₂ (= “Ran”), topic identification information 810D ₃ ( = "Bouncer"), ... are stored in association with the topic identification information 810A.

  In addition, the synonym 900 is associated with the topic identification information 810A. In this example, “works”, “contents”, and “cinema” are stored as synonyms of the keyword “movie” that is the topic identification information 810A. By defining such synonyms, the topic specifying information 810A is obtained when the utterance does not include the keyword “movie” but includes “works”, “contents”, and “cinema” in the utterance sentence or the like. Can be handled as being included in an utterance sentence.

  When the conversation control device 1 according to the present embodiment identifies certain topic identification information by referring to the stored content of the conversation database 500, the other topic identification information stored in association with the topic identification information and It is possible to search and extract topic titles and answer sentences of topic identification information at high speed.

[Topic title]
Next, a data configuration example of the topic title 820 (also referred to as “second morpheme information”) will be described with reference to FIG. FIG. 13 is a diagram illustrating a data configuration example of topic titles.

The topic identification information 810D ₁ , 810D ₂ , 810D ₃ ,... Has a plurality of different topic titles 820 ₁ , 820 ₂ ,..., Topic titles 820 ₃ , 820 ₄ ,..., Topic titles 820 ₅ , 820 ₆ ,. is doing.

In the present embodiment, as shown in FIG. 13, each topic title 820 _{1 to} 820 ₆ ,... Is configured by first specifying information 1301, second specifying information 1302, and third specifying information 1303. Information. Here, in the present embodiment, the first specific information 1301 means main morphemes constituting a topic. As an example of the first specific information, for example, a subject constituting a sentence can be cited. Further, the second specific information 1302 means a morpheme having a close relationship with the first specific information 1301 in the present embodiment. The second specific information 1302 is, for example, an object. Further, in the present embodiment, the third specifying information 1303 means a morpheme that indicates movement of a certain object or a morpheme that modifies a noun or the like. Examples of the third specific information 1303 include verbs, adverbs, and adjectives. The meanings of the first specific information 1301, the second specific information 1302, and the third specific information 1303 do not have to be limited to the above-described contents, and other meanings are the first specific information 1301, the second specific information 1302, Even if the third specific information 1303 is given, the present embodiment is established as long as the contents of the sentence can be grasped from these.

For example, the subject is "Seven Samurai" and the adjective is "interesting", as shown in FIG. 13, the topic title (second morpheme information) 820 ₂ is the first identification information 1301 morpheme " It consists of “Seven Samurai” and the morpheme “Funny” which is the third specific information 1303. Incidentally, this is the topic title 820 ₂ not included morpheme is the second specifying information 1302, the symbol for indicating that there is no corresponding morpheme "*" is stored as the second specification information 1302.

The topic title 820 ₂ (Seven Samurai; *; Interesting) has the meaning of “Seven Samurai is interesting”.
In the present specification, the parentheses constituting the topic title 820 are in the order of the first specific information 1301, the second specific information 1302, and the third specific information 1303 from the left in the following. In addition, in the topic title 820, when there is no morpheme included in the first to third specific information, “*” is indicated for the portion.

  Note that the specific information constituting the topic title 820 is not limited to three like the first to third specific information as described above, and is further limited to other specific information (fourth specific information, and it). You may make it have the above.

[Answer]
Next, the answer sentence 830 will be described. As shown in FIG. 14, in the present embodiment, the reply sentence 830 includes a statement (D; Declaration) and a time (T; Time) in order to make a reply corresponding to the type of utterance sentence uttered by the user. , Location (L; Location), negation (N; Negation), and other types (answer types). The affirmative sentence is “A” and the question sentence is “Q”.
For example, when the topic title (820) 1-1 is (Sato; *; likes) {this is an extracted morpheme included in "I like Sato"}, the topic title (820) 1 The answer sentence (830) 1-1 corresponding to -1 is (DA; statement affirmation sentence "I also like Sato"), (TA; time affirmation sentence "I like Sato when I was standing at bat" ]). An answer acquisition unit 350 described later acquires one answer sentence 830 associated with the topic title 820.

[Conversation control unit]
Here, returning to FIG. 7, a configuration example of the conversation control unit 300 will be described.
The conversation control unit 300 controls the data transfer between the constituent elements (the speech recognition unit 200, the sentence analysis unit 400, the conversation database 500, and the speech recognition dictionary storage unit 600) in the conversation control device 1, and responds to the utterance. Has a function to determine and output an answer sentence.

  In the present embodiment, as shown in FIG. 7, the conversation control unit 300 includes a management unit 310, a topic specifying information search unit 320, an abbreviated sentence complement unit 330, a topic search unit 340, and an answer acquisition unit 350. And have. The management unit 310 operates to control the entire conversation control unit 300 and to pass an answer sentence to the voice output unit 6 as a result of the conversation control process.

The management unit 310 has a function of storing the discourse history and updating it as necessary. In response to requests from the topic identification information search unit 320, the abbreviated sentence complement unit 330, the topic search unit 340, and the answer acquisition unit 350, the management unit 310 converts all or part of the stored discourse history into these units. The function to pass to.
The “discourse history” is information for specifying the topic and subject of the conversation between the user and the conversation control device 1, and the discourse history to be described later is “target topic specification information”, “target topic title”, “user input sentence topic” This information includes at least one of “specific information” and “answer sentence topic specific information”. In addition, “focused topic identification information”, “focused topic title”, and “answer sentence topic specific information” included in the discourse history are not limited to those determined by the previous conversation, but focused topic identification information during a past predetermined period. "Remarked topic title", "Reply sentence topic specific information", or a cumulative record thereof.

Hereinafter, each part which comprises the conversation control part 300 is demonstrated.
[Topic specific information search part]
The topic identification information search unit 320 collates the first morpheme information extracted by the morpheme extraction unit 420 with each topic identification information included in the discourse history, and configures the first morpheme information from the topic identification information. It searches for topic specific information that matches a morpheme. Specifically, the topic identification information search unit 320, when the first morpheme information input from the morpheme extraction unit 420 is composed of two morphemes "Sato" and "like", the input first morpheme The information is collated with the topic specific information group.

  The topic identification information search unit 320 that has performed this collation, when a morpheme constituting the first morpheme information (for example, “Sato”) is included in the focused topic title 820focus (topic title searched up to the previous time), The subject topic title 820focus is output to the answer acquisition unit 350. On the other hand, when the morpheme constituting the first morpheme information is not included in the focused topic title 820focus, the topic identification information search unit 320 determines the user input sentence topic identification information based on the first morpheme information and inputs it. The first morpheme information and the user input sentence topic specifying information are output to the abbreviated sentence complementing unit 330. "User input sentence topic specific information" is included in the topic specific information corresponding to the morpheme corresponding to the content that the user is talking about or the first morpheme information among the morphemes included in the first morpheme information. The topic specific information corresponding to the morpheme which may correspond to the content which the user is talking about among morphemes.

[Abbreviated sentence completion part]
The abbreviated sentence complementing unit 330 uses the first morpheme information as the topic specifying information 810 (hereinafter referred to as “focused topic specifying information”) searched up to the previous time and the topic specifying information 810 (hereinafter referred to as “ A plurality of types of complemented first morpheme information is generated by complementing using “answer sentence topic specifying information”). For example, when the utterance sentence is a sentence “I like”, the abbreviated sentence complementing unit 330 includes the topic topic identification information “Sato” in the first morpheme information “like” and the complemented first morpheme information “ "Sato likes".

  In other words, if the first morpheme information is “W” and the set of the topic topic identification information and the answer sentence topic specification information is “D”, the abbreviated sentence complementing unit 330 adds the set “D” to the first morpheme information “W”. Complemented first morpheme information including elements is generated.

  As a result, when the sentence constructed using the first morpheme information is an abbreviated sentence and is not clear as Japanese, the abbreviated sentence complementing unit 330 uses the set “D” to set the set “D”. (For example, “Sato”) can be included in the first morpheme information “W”. As a result, the abbreviated sentence complementing unit 330 can change the first morpheme information “like” to the first morpheme information “Sato, like”. The complemented first morpheme information “Sato, I like” corresponds to the utterance content “I like Sato”.

  That is, the abbreviated sentence complementing unit 330 can supplement the abbreviated sentence using the set “D” even when the user's utterance content is an abbreviated sentence. As a result, even if the sentence composed of the first morpheme information is an abbreviated sentence, the abbreviated sentence complementing unit 330 can make the sentence in proper Japanese.

  In addition, the abbreviated sentence complementing unit 330 searches for the topic title 820 that matches the first morpheme information after completion based on the set “D”. When a topic title 820 that matches the first morpheme information after complement is found, the abbreviated sentence complement unit 330 outputs the topic title 820 to the answer acquisition unit 350. The answer acquisition unit 350 can output the answer sentence 1030 most suitable for the user's utterance content based on the appropriate topic title 820 searched by the abbreviated sentence complementing unit 330.

  Note that the abbreviated sentence complementing unit 330 is not limited to only including elements of the set “D” in the first morpheme information. The abbreviated sentence complementing unit 330 extracts the first morpheme extracted from the morpheme included in any one of the first specific information, the second specific information, or the third specific information constituting the topic title based on the topic title of interest. It may be included in the information.

[Topic Search Department]
When the topic title 820 is not determined by the abbreviated sentence complementing unit 330, the topic search unit 340 collates the first morpheme information with each topic title 820 corresponding to the user input sentence topic specifying information, and each topic title 820 The topic title 820 that is most suitable for the first morpheme information is searched for.

  Specifically, the topic search unit 340 to which the search command signal is input from the abbreviated sentence complement unit 330 is used based on the user input sentence topic identification information and the first morpheme information included in the input search command signal. The topic title 820 most suitable for the first morpheme information is searched from the topic titles 820 associated with the person input sentence topic identification information. The topic search unit 340 outputs the searched topic title 820 to the answer acquisition unit 350 as a search result signal.

  FIG. 15 is a diagram illustrating a specific example of the topic title 820, the answer sentence 830, the emotion condition parameter 840, and the emotion flag 850 associated with certain topic identification information 810 (= “Sato”). As shown in FIG. 15, for example, the topic search unit 340 includes the topic identification information 810 (= “Sato”) in the input first morpheme information “Sato, I like”, so the topic identification information 810 (= First, the first morpheme information that is input as each topic title (820) 1-1, 1-2,... Associated with the topic specifying information 810 (= “Sato”) Match “Sato, I like”.

  The topic search unit 340, based on the comparison result, the topic title (820) that matches the input first morpheme information “Sato, I like” from among the topic titles (820) 1-1 and 1-2. Specify 1-1 (Sato; *; likes). The topic search unit 340 outputs the searched topic title (820) 1-1 (Sato; *; likes) to the answer acquisition unit 350 as a search result signal. On the other hand, if a topic title that matches the input first morpheme information “Sato, likes” cannot be found, the topic search unit 340 indicates that the user's utterance is an unregistered utterance (response impossible) in the answer acquisition unit 350. To be notified.

[Answer section]
The answer acquisition unit 350 acquires an answer sentence 830, an emotion condition parameter 840, and an emotion flag 850 associated with the topic title 820 based on the topic title 820 searched by the topic search unit 340. Also, the answer acquisition unit 350 collates each answer type associated with the topic title with the utterance type determined by the input type determination unit 440 based on the topic title searched by the topic search unit 340. . The answer acquisition unit 350 that has performed the collation searches for an answer type that matches the determined utterance type from among the answer types.

  As shown in FIG. 15, for example, when the topic title searched by the topic search unit 340 is the topic title 1-1 (Sato; *; likes), the topic title 1-1 The answer type (DA) that matches the “spoken sentence type” (for example, DA) determined by the input type determination unit 440 is identified from the answer sentences 1-1 (DA, TA, etc.) associated with To do.

Further, the answer acquisition unit 350 refers to the emotional state information stored at that time of the emotional state information management unit 7, and selects a plurality of answer sentence candidates corresponding to the specified type of utterance sentence (for example, DA), The emotion condition parameter 840 describing the condition that matches the emotion state information is identified (for example, “all 2 or less”), and the content corresponding to the emotion condition parameter 840 (for example, “I like Sato too”) get.
Here, among the “DA”, “TA”, etc., “A” means an affirmative form. Therefore, when “A” is included in the utterance type and the answer type, it indicates that a certain matter is affirmed. In addition, types such as “DQ” and “TQ” can be included in the utterance type and the answer type. Of these “DQ”, “TQ”, etc., “Q” means a question about a certain matter.

  When the answer type is the question format (Q), the answer sentence 830 associated with the answer type is configured in an affirmative format (A). Examples of the answer sentence 830 created in the affirmative form (A) include a sentence that answers a question item. For example, when the utterance sentence is "Have you operated the slot machine?", The utterance type for this utterance sentence is a question form (Q). The answer sentence 830 associated with this question format (Q) is, for example, “I have operated the slot machine” (affirmative format (A)).

  On the other hand, when the utterance type is an affirmative form (A), the answer sentence associated with the answer type is configured with a question form (Q). Examples of the answer sentence created in this question format (Q) include a question sentence that is replied to the utterance content, or a question sentence that asks a specific matter. For example, when the utterance sentence is “I play with a slot machine”, the utterance type for this utterance sentence is an affirmative form (A). The answer sentence associated with this affirmative form (A) is, for example, “isn't it a hobby to play with pachinko?” (Question sentence (Q) to ask for a specific matter).

  The answer acquisition unit 350 outputs the content of the acquired answer sentence 830 as an answer sentence signal to the management unit 310 and outputs an emotion flag 840 corresponding to the answer sentence 830 to the emotion state information management unit 600. In addition, the management unit 310 to which the answer sentence signal is input from the answer acquisition unit 350 outputs the input answer sentence signal.

  On the other hand, if the user's utterance is an unregistered utterance, the answer acquisition unit 350 requests the unregistered utterance response unit 360 to respond to the unregistered utterance.

[Unregistered utterance response unit]
The unregistered utterance response unit 360 receives the game state information from the game apparatus 2 or receives the state information from the emotion state information management unit 7 when requested to reply to the unregistered utterance from the answer acquisition unit 350 (unregistered utterance information). Emotion state information is received, based on the game state information / emotion state information, one of a plurality of unregistered utterance correspondence response texts prepared as responses to unregistered utterances is selected and passed to the management unit 310.

  Note that the unregistered utterance response answer sentence by the unregistered utterance response unit 360 may be selected based on either or both of the game state information / emotion state information.

  In addition, the unregistered utterance response unit 360 counts the number of times the unregistered utterance is accepted (hereinafter referred to as “unregistered utterance count value”), and selects an unregistered utterance corresponding response sentence according to the unregistered utterance count value. It may be configured to do.

  FIG. 16 is a functional block diagram illustrating a configuration example of the unregistered utterance response unit 360. In the example shown in this figure, the unregistered utterance response unit 360 includes an unregistered utterance correspondence answer sentence selection unit 360A, an unregistered utterance correspondence answer sentence selection unit 360A connected to the unregistered utterance correspondence answer sentence storage unit 360B, An unregistered utterance counter section 360C connected to the registered utterance correspondence answer sentence selection section 360A.

  The unregistered utterance correspondence answer sentence selection unit 360A has a function of selecting an unregistered utterance correspondence answer sentence based on at least one of the game state information, the emotion state information, and the number of times the unregistered utterance is accepted.

  The unregistered utterance correspondence answer sentence storage unit 360B has a function of storing a plurality of unregistered utterance correspondence answer sentences in advance. FIGS. 17A to 17C are unregistered utterance correspondence answer sentence tables stored in the unregistered utterance correspondence answer sentence storage unit 360B, respectively. The unregistered utterance correspondence answer sentence table is an unregistered utterance correspondence answer table. It is a table which memorize | stores the conditions for selecting a sentence, and the content of the answer sentence corresponding to the unregistered utterance corresponding to the condition.

  FIG. 17A shows an example of an unregistered utterance correspondence answer sentence table used when selecting an unregistered utterance correspondence answer sentence based on the game state information. When this unregistered utterance correspondence reply sentence table is used, when the game state information indicates “CPU side initiative” (in the case of the game device 2), when the user utters an unregistered utterance, An unregistered utterance correspondence answer sentence “” is selected and output from the conversation control apparatus 1 as an answer sentence. On the other hand, when the game state information indicates “user side initiative” (when it is the user's turn), when the user utters an unregistered utterance, “Sorry! I did n’t listen. The unregistered utterance correspondence answer sentence “...” Is selected and output from the conversation control apparatus 1 as an answer sentence. When the selection is made using the unregistered utterance correspondence answer sentence table, the answer sentence having the contents corresponding to which number is selected is selected, so that more natural conversation contents can be realized.

  FIG. 17B shows an example of an unregistered utterance correspondence answer sentence table used when an unregistered utterance correspondence answer sentence is selected based on emotion state information. When this unregistered utterance correspondence sentence table is used, when the emotion state information indicating the emotion of the character indicates “good” (in the case of the game device 2), when the user utters an unregistered utterance, An unregistered utterance correspondence answer sentence showing a favorable feeling for the user “Your turn” is selected and output as an answer sentence from the conversation control device 1. On the other hand, when the emotional state information indicates “disgust” (when it is the user's turn), when the user utters an unregistered utterance, it is aggressive against the user who says “Your turn, so do it soon!” An unregistered utterance-response answer sentence with the correct content is selected and output from the conversation control apparatus 1 as an answer sentence. When the selection is made using the unregistered utterance correspondence answer sentence table, an answer sentence corresponding to the emotional state of the character is selected, so that more natural conversation contents can be realized.

  In the above example, the emotional state information is classified into two types of “good” and “disgusted” and the answer sentence corresponding to the unregistered utterance is selected, but this embodiment is limited to this. Instead, conditions other than “good” and “disgust” may be used, and three or more emotion states may be used as conditions for selecting an unregistered utterance correspondence answer sentence instead of two kinds.

  FIG. 17C shows an example of an unregistered utterance correspondence answer sentence table used when selecting an unregistered utterance correspondence answer sentence based on the unregistered utterance count value. When this unregistered utterance correspondence reply sentence table is used, if the unregistered utterance count value is a predetermined number N or more and the user utters an unregistered utterance, the user who is “stubbornly! An unregistered utterance response sentence indicating flicker is selected and output from the conversation control device 1 as an answer sentence. On the other hand, when the unregistered utterance count value is less than the predetermined number N, when the user utters the unregistered utterance, the reply sentence corresponding to the unregistered utterance having a neutral attitude with respect to the user “Sorry, it's your turn”. Is selected and output from the conversation control device 1 as an answer sentence. When selecting using such an unregistered utterance correspondence answer table, if the unregistered utterance is repeated a predetermined number of times, an answer sentence is selected as if the character reacted to the repetition of the unregistered utterance. It becomes possible to realize more natural conversation contents.

Returning to FIG. 16, the description of the unregistered utterance response unit 360 is continued.
The unregistered utterance counter unit 360C has a function of incrementing and storing an unregistered utterance count value every time the unregistered utterance response unit 360 accepts an unregistered utterance. When the unregistered utterance correspondence answer sentence table shown in FIG. 17C is used, the unregistered utterance correspondence answer sentence selection unit 360A refers to the stored contents of the unregistered utterance counter part 360C, and the unregistered utterance correspondence answer Select a sentence. The unregistered utterance counter unit 360C may reset the unregistered utterance count value by using the stored unregistered utterance count value as a trigger when a predetermined period elapses or a predetermined condition is satisfied. An example of the “predetermined period” is “10 minutes”, and an example of the predetermined condition is the end of the game (for example, the end of a half-house).

[Emotional state information management means]
Returning to FIG. 1, the emotion state information management means 7 as a component of the game system GS will be described.
First, a configuration example of the emotional state information management unit 7 will be described with reference to FIG. FIG. 18 is a functional block diagram illustrating a configuration example of the emotion state information management unit 7.

  The emotion state information management means 7 receives an emotion flag 850 that generates a change in the emotion of the character from the conversation control device 1, and an emotion state information update unit 7A that rewrites the emotion state information, and an emotion that stores the emotion state information A state information storage unit 7B and an emotion state information notification unit 7C for notifying all or a part of the emotion state information to the conversation control device 1 and the game device 2 are provided.

  When the emotion state information update unit 7A receives the emotion flag 850, the emotion state information update unit 7A changes the value corresponding to the emotion flag 850 (for example, the count value of the emotion flag) by a predetermined value (for example, +1). Rewrite. When the emotion state information updating unit 7A receives the game state information (for example, information indicating the user's win), the emotion state information updating unit 7A rewrites the emotion state information with a predetermined content (for example, a value corresponding to the emotion flag of “anger”). Or increase the value corresponding to the emotion flag of “joy” or “anger” by twice the current value).

  FIG. 19 shows a data configuration example of emotional state information stored in the emotional state information storage unit 7B. The emotional state information 1900 has a cumulative value 1902 for each emotion flag type 1901. The cumulative value 1902 can be increased or decreased for each emotion flag. For example, every time the emotion state information management means 7 receives an emotion flag indicating “joy”, the value of the cumulative value 1902 is incremented. The emotion state information update unit 7A is stored as a program in advance as to how to increase or decrease, and a predetermined increase / decrease process is executed according to the program.

  Emotion state information notification unit 7C notifies conversation control device 1 and game device 2 of all or part of emotion state information 1900 stored in emotion state information storage unit 7B. The conversation control device 1 refers to the received emotion state information 1900, acquires the content of the answer sentence 830 that satisfies the emotion condition parameter 840 stored in the conversation database 500, and outputs it as a reply to the user. Therefore, this game system GS can output an answer having contents according to the emotion of the character, and can establish a natural conversation. In addition, when the game apparatus 2 receives the emotion state information 1900, the game apparatus 2 performs a character display process so as to indicate the display and operation of the character according to the emotion state information 1900. Thereby, this game system GS can display the character according to the change of the emotion of the character.

[Game system operation example]
An operation example of the game system GS according to the present embodiment having the above-described configuration will be described with reference to FIG. FIG. 20 is a flowchart showing main processing relating to conversation control in the operation example of the game system GS according to the present embodiment. This flowchart does not show processing related to game progress.

  First, the conversation control device 1 accepts an utterance from the user by the voice recognition unit 200, and performs a user utterance acceptance process that changes to data that can be processed by the conversation control unit 300 (step S2001).

Next, the conversation control device 1 performs conversation control processing in order to output an answer sentence corresponding to the utterance from the user (step S2002). The conversation control process (step S2002) can be performed according to the following procedure. FIG. 21 is a flowchart showing the procedure of the conversation control method according to the present embodiment.
First, the voice input unit 3 performs a step of acquiring the utterance content from the user (step S2101). Specifically, the voice input unit 3 acquires the voice constituting the user's utterance content. The voice input unit 3 outputs the acquired voice to the voice recognition unit 200 as a voice signal.

  Next, the voice recognition unit 200 performs a step of specifying a character string corresponding to the utterance content based on the utterance content acquired by the voice input unit 3 (step S2102). Specifically, the voice recognition unit 200 to which a voice signal is input from the voice input unit 3 specifies a word hypothesis (candidate) corresponding to the voice signal based on the input voice signal. The voice recognition unit 200 acquires a character string associated with the identified word hypothesis (candidate), and outputs the acquired character string to the conversation control unit 300 as a character string signal.

  Then, the character string specifying unit 410 performs a step of dividing the series of character strings specified by the voice recognition unit 200 for each sentence (step S2103). Specifically, the character string specifying unit 410 to which a character string signal (or morpheme signal) is input from the management unit 310 has a certain time interval or more in the input series of character strings. , Delimit the string at that part. The character string specifying unit 410 outputs the divided character strings to the morpheme extracting unit 420 and the input type determining unit 440. In addition, when the input character string is a character string input from the keyboard, the character string specifying unit 410 preferably divides the character string at a part such as a punctuation mark or a space.

Thereafter, the morpheme extraction unit 420 performs a step of extracting each morpheme constituting the minimum unit of the character string as the first morpheme information based on the character string specified by the character string specifying unit 410 (step S2104). Specifically, the morpheme extraction unit 420 to which the character string is input from the character string specifying unit 410 collates the input character string with a morpheme group stored in advance in the morpheme database 430. In this embodiment, the morpheme group is prepared as a morpheme dictionary in which each morpheme belonging to each part-of-speech classification describes a morpheme entry word, reading, part-of-speech, utilization form, and the like.
The matched morpheme extraction unit 420 extracts each morpheme (m1, m2,...) That matches each morpheme included in the previously stored morpheme group from the input character string. The morpheme extraction unit 420 outputs each extracted morpheme to the topic identification information search unit 320 as first morpheme information.

  Next, the input type determination unit 440 performs a step of determining “spoken sentence type” based on each morpheme constituting one sentence specified by the character string specifying unit 410 (step S2105). Specifically, the input type determination unit 440, to which the character string is input from the character string specifying unit 410, determines the character string and each dictionary stored in the utterance type database 450 based on the input character string. Collation is performed, and elements related to each dictionary are extracted from the character string. The input type determination unit 440 that extracted this element determines to which “spoken sentence type” the element belongs based on the extracted element. The input type determination unit 440 outputs the determined “spoken sentence type” (speech type) to the answer acquisition unit 350.

Then, the topic specifying information search unit 320 compares the first morpheme information extracted by the morpheme extraction unit 420 with the topic topic title 820focus (step S2106).
If the morpheme constituting the first morpheme information matches the topic topic title 820focus, the topic identification information search unit 320 outputs the topic title 820 to the answer acquisition unit 350. On the other hand, if the morpheme constituting the first morpheme information and the topic title 820 do not match, the topic specific information search unit 320 searches the input first morpheme information and user input sentence topic specific information 810. An abbreviated sentence complementing unit 330 outputs the command signal.

  Thereafter, the abbreviated sentence complementing unit 330 includes the focused topic specifying information and the answer sentence topic specifying information in the input first morpheme information based on the first morpheme information input from the topic specifying information search unit 320. This is performed (step S2107). Specifically, when the first morpheme information is “W” and the set of the topic topic identification information and the answer sentence topic identification information is “D”, the abbreviated sentence complementing unit 330 identifies the topic as the first morpheme information “W”. Complemented first morpheme information including the element of information “D” is generated, and the complemented first morpheme information is collated with all topic titles 820 associated with the set “D” to be complemented. Whether there is a topic title 820 that matches the first morpheme information is searched. If there is a topic title 820 that matches the complemented first morpheme information, the abbreviated sentence complementing unit 330 outputs the topic title 820 to the answer acquisition unit 350. On the other hand, when the topic title 820 that matches the supplemented first morpheme information is not found, the abbreviated sentence complementing unit 330 passes the first morpheme information and the user input sentence topic specifying information to the topic search unit 340. .

  Next, the topic search unit 340 collates the first morpheme information with the user input sentence topic identification information, and performs a step of searching for the topic title 820 suitable for the first morpheme information from each topic title 820. (Step S2108). Specifically, the topic search unit 340 to which the search command signal is input from the abbreviated sentence complement unit 330 is based on the user input sentence topic identification information and the first morpheme information included in the input search command signal. A topic title 820 suitable for the first morpheme information is searched from the topic titles 820 associated with the user input sentence topic identification information. The topic search unit 340 outputs the topic title 820 obtained as a result of the search to the answer acquisition unit 350 as a search result signal.

  Next, the answer acquisition unit 350 determines the utterance type and the topic title 820 of the user determined based on the topic title 820 searched by the topic identification information search unit 320, the abbreviated sentence complement unit 330, or the topic search unit 340. Collate each associated answer type. The answer acquisition unit 350 that has performed this collation searches the answer types that match the determined utterance type from among the answer types (step S2109).

  Specifically, the answer acquisition unit 350 to which the search result signal is input from the topic search unit 340 and the “spoken sentence type” is input from the input type determination unit 440 is the “topic title” corresponding to the input search result signal. Based on the entered “spoken sentence type”, the answer type matching the “spoken sentence type” (such as DA) is selected from the answer type group associated with the “topic title”. Identify.

  The answer acquisition unit 350 acquires an answer sentence 830 associated with the specified answer type from the conversation database 500. The answer acquisition unit 350 outputs the acquired answer sentence 830 to the output unit 600 via the management unit 310. The output unit 600 to which the answer sentence is input from the management unit 310 outputs the input answer sentence 830 (step S2110).

  Returning to FIG. 20, the description of the operation example of the game system GS is continued. In the conversation control process (step S2002), when the topic identification information corresponding to the user's utterance is not in the conversation database 500 (YES in step S2003), the unregistered utterance response unit 360 of the conversation control unit 300 is in the game state. Conversation control is performed by selecting and outputting an unregistered utterance correspondence answer sentence in accordance with emotion state information, an unsupported utterance count value, and the like (step S2004).

  On the other hand, in the conversation control process (step S2002), when the topic identification information corresponding to the user's utterance is in the conversation database 500 (step S2003, NO), the answer text is selected and output in step S2002. Therefore, no special processing is performed.

Next, the game system GS determines whether the game end condition is satisfied (step S2005). When the game end condition is satisfied (for example, the end of the game time), the game system GS ends the game. On the other hand, when the game end condition is not satisfied, the process returns to step S2001 again and waits for acceptance of the user utterance.
This completes the main processing of the game system (except for processing related to game control).
[Other variations]
(1) The configurations and operations of the speech recognition unit 200, the conversation control unit 300, and the sentence analysis unit 400 of the conversation control device 1 are not limited to those described in the above embodiment, and the user can use the conversation database means. Any speech recognition unit 200, conversation control unit 300, and sentence analysis unit 400 other than those described in the embodiment as long as it returns an answer corresponding to the utterance content, the conversation control according to the present invention. It can be used as a component of the device 1.
(2) In the description of the above embodiment, the input of the utterance to the conversation control device 1 has been described as being by voice, but the input of the utterance to the conversation control device 1 is not limited to voice, but a keyboard, a touch panel An utterance is input to the conversation control device 1 as character string data by the non-voice input means 4 such as a pointing device, and the conversation control device 1 answers the utterance input as the character string data using the conversation database 500. The present invention is also established as a configuration for outputting a sentence.

Block diagram showing a configuration example of a game system Functional block diagram showing a configuration example of a game device Functional block diagram showing a configuration example of a conversation control device Functional block diagram showing a configuration example of the speech recognition unit Timing chart showing an example of processing of the word hypothesis narrowing unit Flow diagram showing an example of the operation of the voice recognition unit Partial enlarged block diagram of the conversation control device The figure which shows the relationship between the character string and the morpheme extracted from this character string The figure which shows the example of the utterance sentence which corresponds to the type of the utterance sentence, the two letter alphabet which shows the type of the utterance sentence, and the type of the utterance sentence The figure which shows the relationship between the type of sentence and the dictionary for judging the type Conceptual diagram showing an example of the data structure of data stored in the conversation database The figure which shows the correlation with a certain topic specific information and other topic specific information Diagram showing data structure example of topic title Illustration explaining the type of answer text The figure which shows the specific example of the topic title matched with a certain topic specific information, an answer sentence, an emotion condition parameter, and an emotion flag Functional block diagram showing a configuration example of an unregistered utterance response unit The figure which shows the data structural example of the unregistered utterance corresponding reply sentence table memorize | stored in the unregistered utterance corresponding reply sentence memory | storage part Functional block diagram showing a configuration example of emotion state information management means The figure which shows the data structural example of the emotion state information memorize | stored in an emotion state information storage part Flow chart showing the main operations of the game system Flow chart showing an example of conversation control processing

Explanation of symbols

GS ... Game system 1 ... Conversation control device 2 ... Game device 3 ... Voice input means 4 ... Non-voice input means 5 ... Image output means 6 ... Voice output means 7 ... Emotion state information management means 200 ... Voice recognition unit 300 ... Conversation control Unit 360 ... unregistered utterance response unit 400 ... sentence analysis unit 500 ... conversation database 600 ... speech recognition dictionary storage unit

Claims (12)

In a game machine having an input means for receiving an utterance from a user, a conversation processing means for outputting an answer sentence in response to the utterance from the user, and a game control means for controlling a character and executing a game,
The conversation processing means includes
A recognition unit that receives and recognizes the utterance from the user accepted by the input means as the utterance information from the user;
Means for storing a discourse history that is information for identifying the topic and subject of conversation between the user and the conversation processing means;
Conversation storing a plurality of topic specifying information determined by the discourse history and associated with a topic title composed of a morpheme composed of one character, a plurality of character strings, or a combination thereof and an answer sentence to the user Database means;
A morpheme extraction unit that extracts speech information from the user into a morpheme composed of one character, a plurality of character strings, or a combination thereof;
Collating the morpheme information extracted by the morpheme extraction unit with the topic specifying information stored in the conversation database means, and selecting the topic specifying information that matches the morpheme constituting the morpheme information from each topic specifying information Means for selecting the answer sentence to the user by:
When the topic identification information corresponding to the user's utterance is unregistered in the conversation database means, it is not possible to find the topic identification information that matches the morpheme constituting the morpheme information from each topic identification information. An unregistered utterance response unit that receives unregistered utterance information to inform the effect,
With
The unregistered utterance response unit
An unregistered utterance correspondence answer sentence storage unit for storing a plurality of unregistered utterance correspondence answer sentences;
When the unregistered utterance information is received, a predetermined one of the unregistered utterance correspondence answer sentences stored in the unregistered utterance correspondence answer sentence storage unit based on the game state information regarding the progress state of the game output from the game control means. An unregistered utterance correspondence answer sentence selection unit for selecting an unregistered utterance correspondence answer sentence,
A game machine comprising: An emotion flag constituting emotion state information indicating the emotion of the character is associated with the topic title, and the game machine receives an emotion flag associated with the selected answer sentence and based on the emotion flag Emotional state information management means for updating emotional state information is further provided,
When the unregistered utterance correspondence answer sentence selection unit receives the unregistered utterance information, the unregistered utterance correspondence answer sentence storage unit stores the unregistered utterance correspondence answer sentence storage unit based on the emotion state information output from the emotion state information management unit. The game machine according to claim 1 , wherein a predetermined unregistered utterance correspondence answer sentence is selected from the utterance correspondence answer sentences . The unregistered utterance response unit further includes an unregistered utterance counter unit that increments and stores an unregistered utterance count value each time the unregistered utterance information is received,
When the unregistered utterance correspondence answer sentence selection unit receives the unregistered utterance information, the unregistered utterance correspondence answer sentence storage unit stores the unregistered utterance correspondence answer sentence storage unit based on the unregistered utterance count value output from the unregistered utterance counter unit. The game machine according to claim 1 or 2 , wherein a predetermined unregistered utterance correspondence answer sentence is selected from unregistered utterance correspondence answer sentences . Along with recognizing speech information as a series of character strings, a character string specifying unit that separates the character strings into one sentence,
An input type determining unit that determines the type of utterance content based on the character string specified by the character string specifying unit, and outputs information indicating the utterance type to the means for selecting the answer sentence according to the determination result;
The game machine according to any one of claims 1 to 3, further comprising: It further comprises an abbreviated sentence complementing unit that generates a plurality of types of complemented morpheme information by complementing the morpheme information using the topic identification information included in the previous topic identification information and the previous answer sentence. The game machine according to any one of claims 1 to 4 . An input means for receiving an utterance from the user; a conversation processing means for outputting an answer sentence in response to the utterance from the user; and a game control means for controlling a character and executing a game, wherein the conversation processing means A means for storing a discourse history, which is information for specifying a topic or subject of conversation between the user and the conversation processing means, and a morpheme composed of one character, a plurality of character strings, or a combination thereof, and In a method of executing a game on a game machine comprising a conversation database means for storing a plurality of topic identification information determined by the discourse history and a topic title associated with the answer sentence of
The recognition unit of the conversation processing means receives and recognizes the utterance from the user accepted by the input means as the utterance information from the user; and
The morpheme extraction unit of the conversation processing means extracts the utterance information from the user into a morpheme consisting of one character, a plurality of character strings, or a combination thereof;
The topic identification information search unit of the conversation processing unit collates the morpheme information extracted by the morpheme extraction unit with the topic identification information stored in the conversation database unit, and constructs morpheme information from each topic identification information Searching for topic specific information that matches a morpheme to
When the topic identification information corresponding to the user's utterance is registered in the conversation database unit, the answer acquisition unit of the conversation processing unit is associated with the topic identification information searched by the topic identification information search unit Retrieving a sentence;
When the topic identification information corresponding to the user's utterance is not registered in the conversation database means, the topic identification information search unit finds the topic identification information that matches the morpheme constituting the morpheme information from each topic identification information. The unregistered utterance response unit of the conversation processing means receives from the response acquisition unit unregistered utterance information informing that, if not possible,
Including
When the unregistered utterance response unit receives the unregistered utterance information, the unregistered utterance response unit stores the unregistered utterance correspondence answer sentence storage unit based on the game state information regarding the progress of the game output from the game control unit. A method comprising: selecting and outputting a predetermined unregistered utterance correspondence answer sentence from utterance correspondence answer sentences . The emotion flag constituting emotion state information indicating the emotion of the character is associated with the topic title, and the game machine receives the emotion flag associated with the answer sentence acquired by the answer acquisition unit and the emotion It further comprises emotion state information management means for updating the emotion state information based on the flag,
  When the unregistered utterance correspondence answer sentence selection unit receives the unregistered utterance information, the unregistered utterance correspondence answer sentence storage unit stores the unregistered utterance correspondence answer sentence storage unit based on the emotion state information output from the emotion state information management unit. The method according to claim 6, wherein a predetermined unregistered utterance correspondence answer sentence is selected from the utterance correspondence answer sentences. The unregistered utterance response unit further includes an unregistered utterance counter unit that increments and stores an unregistered utterance count value each time the unregistered utterance information is received,
  When the unregistered utterance correspondence answer sentence selection unit receives the unregistered utterance information, the unregistered utterance correspondence answer sentence storage unit stores the unregistered utterance correspondence answer sentence storage unit based on the unregistered utterance count value output from the unregistered utterance counter unit. The method according to claim 6 or 7, wherein a predetermined unregistered utterance correspondence answer sentence is selected from unregistered utterance correspondence answer sentences. After the step in which the morpheme extraction unit of the conversation processing unit extracts the utterance information from the user into the morpheme, the character string specifying unit of the conversation processing unit recognizes the utterance information as a series of character strings, and the character string A step for separating each sentence into phrases,
The input type determination unit of the conversation processing means determines the type of utterance content based on the character string specified by the character string specifying unit, and outputs information indicating the utterance type to the answer acquisition unit according to the determination result And steps to
The method according to claim 6, further comprising: After the step in which the topic specifying information search unit of the conversation processing means searches for the topic specifying information, the abbreviated sentence complementing unit of the conversation processing means obtains the topic specifying information included in the previous topic specifying information and the previous answer sentence. The method according to any one of claims 6 to 9, further comprising generating a plurality of types of complemented morpheme information by complementing the morpheme information using the method . The program for functioning a computer as a game machine of any one of Claims 1 thru | or 5 . The program for making a computer perform the method of any one of Claim 6 thru | or 10 .

Images