JP2017152948A - Information provision method, information provision program, and information provision system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To create a method for providing a user with optimum information on a service of a company by creating a new point of contact between the company and the user via a communication network.SOLUTION: An information provision method from a company to a user includes: displaying a virtual operator on a communication terminal of the user; and, via an interaction between the user and the virtual operator, selecting an optimum answer from a plurality of answers corresponding to a request of the user to provide the selected answer to the user. The interaction is executed by the user inputting voice information via the communication terminal, transmitting the voice information to a voice recognition server via a network, the voice recognition server recognizing the voice information, an interaction server understanding an intention of the recognized voice information, selecting an answer message corresponding to the intention understood by the interaction server from a database, a voice synthesis server outputting the answer message, and outputting the output answer message as voice of the virtual operator displayed on the communication terminal.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an information providing method from a business operator to a customer or user, and a program and system for realizing the information providing method.

  With the widespread use of communication terminals such as computers and mobile phones, and the rapid development of communication networks, businesses and customers or users who provide various services (hereinafter referred to as users in this specification and claims) The form of business between the user and the user as a whole is also changing greatly. For example, in banks and securities companies that provide financial products as services, users can purchase services provided by business operators by visiting the store, or by making a verbal call directly to the business operator over the telephone, etc. Various problems related to the problem were solved. Nowadays, with the development of communication networks, users have been able to receive services via the communication network and solve problems. A typical example is Internet banking, and a user can perform various commercial transactions by operating a communication terminal regardless of time and place. Such changes in commerce are the same not only in the financial industry but also in various business forms.

  Recently, speech recognition technology equipped with artificial intelligence has been developed, and a support system has been developed for work that has been directly handled by the operator or its operator, such as handling requests from users and handling complaints. For example, in Patent Document 1, the user inputs voice information to the communication terminal, and the user's request, question, and complaint intention are extracted based on the analysis result of the input voice information. A system that supports quick response is disclosed.

JP 05-216493 A

  One of the problems of the present invention is to create a new point of contact between a business operator and a user through a communication network, and to create a method for providing the user with optimal information related to the business of the business by using this. It is. Alternatively, one of the problems of the present invention is to build a more intimate user-business relationship by reducing the sense of distance between the user and the business through means such as a natural phone call via a mobile terminal, so that the user can Giving you the opportunity to enjoy the experience. Alternatively, one of the problems of the present invention is to reduce the business burden on the business operator by avoiding the consumption of human resources due to user correspondence, and to limit the abundant information to the knowledge, experience, and ability of the business operator. It is to create a method of providing a high-quality and high-value-added service to the user in a timely manner by providing it to the user. Alternatively, one of the objects of the present invention is to provide a program and a system for realizing the above method.

  One embodiment of the present invention is an information providing system. The information providing system includes a voice recognition unit that recognizes user voice information transmitted from a communication terminal, a dialog unit that selects a response message for the voice information from a database in which a plurality of messages are stored, and a selected answer. A transmitting unit that transmits the message to the communication terminal as at least one of voice information and character information, and the dialog unit displays an inquiry message for asking a user a question when the answer message cannot be narrowed down to one Select from the database and send the selected inquiry message to the communication terminal.

  The information providing system may further include a communication terminal. The communication terminal can be configured to display a virtual operator. In addition, the information providing system may be configured such that when the answer message and the inquiry message are reproduced on the communication terminal, a part of the virtual operator performs a large operation compared to when the answer message and the inquiry message are not reproduced. it can.

  The information providing system further includes a first server in which standard biometric information of the user is stored, and the communication terminal includes an image capturing device that captures the biometric information of the user, based on the biometric information and the standard biometric information. It can be configured to allow the user to input voice information when the user is approved.

  The information providing system may further include a second server for calling between the user and the operator.

  The information providing system can be further configured such that the second server stores the voice information recognized by the voice recognition unit, the selected answer message, and the inquiry message.

  The information providing system further includes a third server in which map information is stored, the first server is configured to store information on a plurality of stores of the business operator, and the position of the communication terminal and the plurality of information are displayed on the display unit. The communication terminal can be configured to display the store information.

  The dialogue unit can be configured to select at least one store from a plurality of stores based on the audio information, and the third server can be configured to transmit the route to the selected store to the communication terminal.

  In one embodiment of the present invention, an answer message corresponding to a question inputted by voice by a user is narrowed down from a plurality of messages, and when the answer message cannot be narrowed down to one, an inquiry message for asking a question to the user Is a server configured to select from a plurality of messages.

  The server can be configured such that the question is transmitted from a voice recognition server that recognizes voice information input by the user.

  The server can be configured such that the answer message and the inquiry message are transmitted to the speech synthesis server.

  The server may be connected to a first server configured to store user biometric information.

  The first server can be configured to store business store information.

  The first server may be connected to a second server configured to store at least one of a user question, a selected answer message, and a selected query message.

  One embodiment of the present invention recognizes voice information including a user's question transmitted from a communication terminal, selects an answer message corresponding to the recognized user's question from a plurality of answer messages, and is selected. The response message is transmitted to the communication terminal as at least one of voice information and character information, and when the response message cannot be narrowed down to one, the inquiry message for asking the user a question is selected from a plurality of response messages. , A program for executing transmission of a selected inquiry message to a communication terminal.

  The program can be configured to receive user biometric information captured by the communication terminal and execute user authentication using the biometric information.

  The program may be further configured to store at least one of a user question, a selected answer message, and a selected inquiry message.

  The program is configured to refer to at least one of the stored user question, the selected answer message, and the selected inquiry message, and to further connect the communication terminal and the telephone terminal of the operator. can do.

  The program refers to at least one of the stored user question, the selected answer message, and the selected inquiry message, and is configured to further execute displaying the store information of the operator on the communication terminal. be able to.

  The program can be configured to further execute receiving a question as character information from the communication terminal and transmitting an answer message as character information to the communication terminal.

  The program can be configured to further execute receiving a question as text information from the communication terminal and transmitting an answer message to the communication terminal as text information when a question included in the voice information cannot be detected. .

  The program can be configured to receive volume data around the communication terminal from the communication terminal and to receive a question as character information from the communication terminal according to the volume data.

  One embodiment of the present invention transmits speech information to a system including a speech recognition server, a dialogue server, and a speech synthesis server, and at least one of an answer message for a user question and an inquiry message for a question included in the speech information. Is a program that causes a communication terminal to execute a response message and an inquiry message as at least one of voice information and character information.

  The program can be configured to be stored in a communication terminal.

  The program can be configured to be acquired from a communication network.

  The program can be configured to display a virtual operator. The program is configured such that when the answer message and the inquiry message are reproduced on the communication terminal, a part of the virtual operator performs a large operation compared to when the answer message and the inquiry message are not reproduced. can do.

  The said program can be comprised so that a user's biometric information may be acquired.

  The program can be configured to connect the communication terminal and the telephone terminal of the business operator.

  The said program can be comprised so that the positional information on a communication terminal may be acquired, a map may be displayed on a communication terminal, and a shop's shop information may be displayed on a communication terminal.

  The program can be configured to display a route from the position of the communication terminal to the business operator's store.

  The program can be configured to further execute sending the question as character information.

  The program can be configured to measure the volume around the communication terminal and to transmit the volume data to at least one of a voice recognition server, a dialogue server, and a voice synthesis server.

  The program can be configured to further execute transmitting the question as character information in accordance with the volume.

  In one embodiment of the present invention, voice information input by a user is transmitted to a communication terminal, the transmitted voice information is recognized, and a plurality of answer messages corresponding to user questions included in the recognized voice information are received. Inquiry to ask the user a question if the selected response message is transmitted to the communication terminal as at least one of voice information and character information and the response message cannot be narrowed down to one. This is an information providing method in which a message is selected from a plurality of reply messages and the selected inquiry message is transmitted to a communication terminal.

  The information providing method may further include displaying a virtual operator on the communication terminal. In addition, when the answer message and the inquiry message are reproduced on the communication terminal, a part of the virtual operator may be largely moved as compared with a case where the answer message and the inquiry message are not reproduced.

  The information providing method may further include transmitting user biometric information captured by the communication terminal and authenticating the user using the transmitted biometric information.

  The information providing method may further include storing at least one of a user question, a selected answer message, and a selected inquiry message.

  The information providing method may further include referring to at least one of the stored user question, the selected answer message, and the selected inquiry message, and connecting the communication terminal and the operator's telephone terminal. it can.

  The information providing method may include displaying at least one of the stored user question, the selected answer message, and the selected inquiry message, and displaying the store information of the operator on the communication terminal.

  The information providing method includes inputting a question as character information (text) to a communication terminal, transmitting the question, transmitting an answer message as character information to the communication terminal, and displaying the answer message on the communication terminal. Can do.

  The information providing method may include measuring surrounding sound volume using a communication terminal. Also, depending on the volume, it may include inputting the question as character information to the communication terminal, transmitting the question, transmitting the answer message to the communication terminal as character information, and displaying the answer message on the communication terminal.

  By having the user and the virtual operator interact, an intimate user-business relationship can be established. In addition, through the interaction between the user and the virtual operator, it is possible to find out information desired by the user from among a lot of information that can be provided by the business operator. In other words, the user can be “guided and guided” to the required information through dialogue without forcing the user to perform the complicated task of “selecting” from the large amount of information provided by the operator. A user experience can be provided. In addition, by avoiding the consumption of human resources in response to users, the business efficiency of the business operator and the improvement of productivity can be achieved, and uniform and abundant information can be used for the knowledge, experience and ability of the business operator and its operators. By providing the user without limitation, a service having high quality and high added value can be provided in a timely manner. Further, the user can easily grasp the amount of resources managed by the business operator and the variation thereof.

The block diagram of the communication terminal applied to the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention. The structural example of the information provision system of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. An example of the image displayed in the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention. The flowchart explaining the information provision method of embodiment of this invention.

  Hereinafter, an information providing method according to an embodiment of the present invention, and a program and a system for realizing the method will be described in detail with reference to the drawings. The following embodiments are examples of the embodiments of the present invention, and the present invention is not limited to these embodiments.

<First Embodiment>
An information providing method according to the first embodiment of the present invention will be described with reference to the drawings. In the present embodiment, the user performs a natural dialogue with an image imitating an operator of a business operator (displayed as a virtual operator in the present specification and claims) displayed on the communication terminal using the communication terminal, In this manner, an information providing method for a provider to provide answers and information suitable for the user, and a program for executing the information will be described.

[1. Preparation]
FIG. 1 shows a system diagram of a communication terminal 100 that can be used in the information providing method of the present embodiment, and FIG. 2 shows a flowchart of the information providing method of the present embodiment. First, the user prepares the communication terminal 100 in which an application including a program provided by the operator is installed. The application may be provided in a state stored in a medium, or may be distributed via a communication network and acquired by a user.

  The communication terminal 100 is a terminal that can be operated by the user, and includes a stationary (desktop) computer, a notebook computer, or a portable communication terminal such as a mobile phone, a tablet, or a smartphone. A configuration example of the communication terminal 100 is shown in FIG. As shown in FIG. 2, the communication terminal 100 includes a CPU (Central Processing Unit) 102, a memory 104, a display unit 106, an input unit 108, an audio input unit 110, an audio output unit 112, a communication unit 114, and an image capturing device 116. Etc. are included.

  The CPU 102 has a function of interpreting and executing various programs stored in the communication terminal 100. The memory 104 has a function of storing information such as programs and data, and includes a nonvolatile memory element such as a flash memory. The display unit 106 includes a liquid crystal display device, an EL organic device, and the like, and functions as a user interface. The input unit 108 has a function as an interface through which a user inputs information to the communication terminal through physical contact, and typically includes a touch panel, a keyboard, an input button, and the like. The voice input unit 110 has a function of capturing voice and transmitting the input voice information to the communication unit 114. The audio output unit 112 has a function of reproducing audio information transmitted from a server and audio information transmitted from another communication terminal through a communication network such as a telephone line. The communication unit 114 has a function of transmitting and receiving voice information, character information, and the like with a server and other communication terminals through a communication network.

  Here, the server in this specification and claims may be a server that is maintained, managed, and used by the operator itself, or a server that is maintained, managed, and provided by an information processing service provider that is different from the provider. There may be. As shown below, many of the systems for realizing the information providing method of the present embodiment can use a server managed and provided by an information processing service provider. Information that cannot be disclosed by the operator, for example, information related to the user's personal information, is preferably stored and used in a server that is maintained and managed by the operator.

[2. Start-up]
Referring to FIG. 2, when the user starts an application (step 200), the program displays a virtual operator on the display unit 106 of the communication terminal (step 202). FIG. 3 is an example of a virtual operator, and illustrates a state displayed on the display unit 106 of the communication terminal 100. The virtual operator may be a model of an employee who handles a user (hereinafter referred to as an operator). Further, the program may be configured so that the virtual operator performs natural human actions (for example, changes in facial expressions, head and hand movements, etc.) and makes the user feel close.

  As described above, FIG. 3 shows a virtual operator imitating an operator, but there is no restriction on the form of the virtual operator. The program may be configured so that the user can arbitrarily select and set the gender and appearance of the virtual operator, the clothes, or the background of the virtual operator (step 204). Thus, the user can not only select a virtual operator based on personal preference, but also can fix a virtual operator dedicated to the user. By giving such a function, it is possible to make the user have a so-called “customer” consciousness and contribute to the establishment of a more intimate user-business relationship.

  Here, the program may be configured to execute an authentication process for the user as necessary. For example, if the content of a user request, inquiry, or question (hereinafter referred to as a question in the present specification and claims) is required to have a high level of confidentiality, an authentication process (step 206) is executed. It is preferable.

  In the authentication process (step 206), for example, the image capturing device 116 provided in the communication terminal 100 is used to capture biometric information such as an image of a user's face, a fingerprint or a vein pattern, and the user's information stored in the server in advance. Comparison with standard data (standard biological information) can be performed based on the result. Alternatively, the authentication process may be performed by requesting the user to input a password and comparing the password with the user password stored in the server. Alternatively, user authentication may be requested and user authentication may be performed using the input voice (voice print). In the authentication process, if it is determined that the communication terminal 100 is operated by a person other than the user, the program may be configured to automatically notify the operator.

  When it is determined that the user authentication is normally completed and the communication terminal is operated by the user, the service content provided by the operator in the application may be changed for each user. Alternatively, an answer message to be notified to the user may be output as voice or displayed on the display unit 106 as characters (step 208).

[3. basic action]

  As illustrated in FIG. 3, the information providing method and the program according to the present embodiment are configured to display, for example, an icon 300 in addition to the virtual operator. The icon 300 is an icon that activates the voice input unit 110, and the user can start voice input by operating the icon 300. Moreover, the program may be comprised so that the icon 302 for shifting to the mode which displays a shop information of a provider may be displayed. As will be described later, by using this icon 302, the program may be configured to display, for example, the service contents provided by each store of the business operator, the congestion status and position of the store, the route to the store, and the like.

  First, the user operates the icon 300 to activate the voice input unit 110. Thereafter, the user verbally asks questions about the business (step 210). In accordance with the instructions of the program, the voice input unit 110 transmits the user's question as voice information to the communication unit 114, and the communication unit 114 passes the communication network to the API (application programming interface) serving as an interface of the voice recognition server. Is transmitted (step 212). The transmitted voice information is analyzed in the voice recognition server and converted into character data (text data) (step 214).

  The input of the question is not limited to voice input, and may be performed by inputting character information using the input unit 108 of the communication terminal 100. For example, character input (text input) is effective when utterance is prohibited or when an application is to be used in a frivolous environment. In such a case, it is possible to shift to an operation mode (web chat mode) based only on character input and output as will be described later.

  The voice information converted into the text data is analyzed by the dialogue server, and the intention of the user's question is understood and grasped (step 216). If necessary, the dialogue server selects a message (confirmation message) for confirming whether or not the intention is correctly understood from a server in which various messages are stored or a database in the server (step 218). Thereafter, the character information of the confirmation message is transmitted to the voice synthesis server, and the voice synthesis server converts the character information into voice information, and transmits the voice information to the user communication terminal 100 through the API of the voice synthesis server (step 218). . Thereafter, the audio output unit 112 reproduces the audio information received by the communication terminal 100 as audio in accordance with the instructions of the program (step 220). The confirmation message may be displayed on the display unit 106 as character information, and may be reproduced as voice at the same time. Since voice information is generated by the voice synthesis server, it is possible to realize a voice uttered by an actual person.

  If the user recognizes that the dialog server correctly understands the intention, the user touches the icon 300 to activate the voice input unit 110 and inputs the voice information to the voice input unit 110 as voice information. Can be used to enter text (step 222). On the other hand, when the dialog server recognizes that the intention is not correctly understood, the voice input unit 110 is activated by touching the icon 300 and the question is input to the voice input unit 110 again, or the input unit 108 is operated to enter the text. The question may be input again in the form (step 222).

  After understanding the intention of the user's question, the dialogue server narrows down the answer messages using a narrowing engine (search engine) or the like (step 224). Then, a suitable answer message is selected from various answer messages stored in the database in the dialog server (step 226). The character information of the selected answer message is converted into speech information by the speech synthesis server and transmitted to the user communication terminal 100 through the API of the speech synthesis server (step 226). At this time, the reply message may be transmitted to the communication terminal 100 as text information together with the voice information. The voice output unit 112 reproduces the voice information received by the communication terminal 100 in accordance with the instructions of the program (step 228). At the same time, the answer message may be displayed as text on the display unit 106.

  A fixed message created by the operator may be stored as a reply message in the database, and the dialog server can select a reply message from the stored messages. However, the information providing method according to the present embodiment is not limited to this. In addition to selecting an answer message, a function as an artificial intelligence is installed in the dialogue server, and a new answer message is created by learning through dialogue with the user. This may be transmitted to the communication terminal 100 after being converted into voice information by the voice synthesis server. That is, a function for acquiring a new conclusion from a dialog with the user and a function for learning from a dialog history with the user may be given to the dialog server, and an answer message may be created using these functions.

  For example, when a final reply message satisfying the user is obtained through the dialogue between the user and the virtual operator, the series of dialogue at this time is stored as one component. The voice information of the user included in this component is morphologically analyzed, and the keyword that led to the final response message is extracted. At the same time, the keywords are weighted and the relevance (grammatical relevance and positional relevance) between the keywords is parameterized, and this information is finally assigned to the selected answer message. By accumulating such processes, the keyword and the relevance that the user who ultimately needs each answer message will give are learned. Therefore, when a new conversation is started, the keyword extracted from the user's voice information and its relation are analyzed, and the most suitable answer message selected in the past based on the analysis result is selected. Can do. As a result, the same optimal answer can always be selected for similar or similar questions. In addition, a plurality of response messages can be merged by the speech synthesis server as needed to generate a new message and provide it to the user.

  If the user determines that a satisfactory answer to the question has been obtained from the reproduced voice information or the displayed character information, the user can end the application (step 230). On the other hand, if the user determines that a sufficient answer has not been obtained, the user can input the same question or a similar question again by voice input or text input to obtain an answer (step 232). ). The subsequent processes are the same as those described above.

  In step 224, if the answer message for the user's question cannot be narrowed down or if a large number of answer messages must be selected, a question (inquiry message) for the user is generated or selected and transmitted to the communication terminal 100; The user may be requested to provide further information (step 234). The inquiry message is reproduced and displayed on the communication terminal 100 as voice, text, or both. The user can answer the inquiry message by voice or text (step 222). By repeating the steps of sending the inquiry message and answering it, it becomes possible to find out the answer and information desired by the user from among a lot of information that can be provided by the operator through the dialogue between the user and the virtual operator. Therefore, the information providing method according to the present embodiment of the present invention does not require the user to perform a complicated task of “selecting” from a lot of information provided by the operator, and through the dialogue, It can “guide and guide” users to information in real time and provide a new user experience.

  Although not shown in FIG. 2, a message for chatting with the user may be stored in the database of the dialogue server, and the program may be configured so that the user can chat with the virtual operator. . As a result, the tension of the user can be relieved and a more intimate user-business relationship can be established. Furthermore, the program may be configured to display an icon for shifting to a site capable of real-time communication (chat) on the communication network. This makes it possible to quickly share, for example, useful information provided by the operator with a third party, which not only improves the service for the user, but also allows the operator to obtain an effective advertising effect. it can.

  Further, when the various messages described above are reproduced by the voice output unit 112, the virtual operator's mouth and eyes may be moved in order to make it appear that the virtual operator is actually speaking. Or you may make it carry out the gesture which explains, such as moving a hand. That is, the program may be configured such that a part of the virtual operator performs a larger operation compared to when the message is not reproduced.

  As described above, in the present embodiment, the user can communicate various questions to the operator through the dialogue between the virtual operator and the user. Therefore, as long as the usage environment permits, the user is freed from the character input operation using the keyboard and the touch panel, and can easily transmit various questions to the business operator. In addition, since the user interacts with the virtual operator using the speech synthesis technology, the user can perform a natural conversation without being forced to perform an inorganic conversation. On the other hand, users get answers through dialogues with virtual operators, not dialogues with operators or their operators, so users can ask various questions without worrying more than necessary, creating new user experiences. You can enjoy it.

  In the steps described above, the operator of the business operator is not directly involved in the dialog with the user. Therefore, the user can acquire uniform and abundant information in real time without being influenced by the knowledge, experience, and ability of the operator. In addition, since the operator does not need human resources at this stage, it is possible to avoid the consumption of human resources in correspondence with the user, and it is possible to improve the efficiency of business and the improvement of productivity.

[4. Switch to call mode]
Answers frequently asked by many users (FAQ) can be prepared in advance, and high user satisfaction can be obtained. However, there are cases where the intention of the user's question cannot be dealt with only by the message data stored in the database in the dialogue server. For example, in the case where a question other than FAQ, complicated consultation contents, or highly confidential information is included in the question, the method described above may not be available. In such a case, it is possible to switch to a mode in which a telephone call is used to directly talk to the operator or its operator.

  For example, as shown in FIG. 4, after the reply message from the server is reproduced as voice information (steps 224 and 226), an icon (telephone icon, not shown in FIG. 3) for switching to the call mode is displayed on the communication terminal 100. The program may be configured to display on the display unit 106 (step 404). The communication terminal 100 is connected to the communication service cooperation server via the API when the user operates the telephone icon according to the command of the program. The communication service cooperation server connects the telephone from the user to the communication network, and enables a telephone call with the operator or the operator via the telephone line (step 406).

  The telephone icon for switching to the call mode may be displayed on the display unit 106 from the application activation stage. However, depending on the operator's human resources, it is necessary to restrict the user's transition to the call mode to some extent. In this case, the telephone icon is not displayed when the application is started, and the dialog server selects and transmits an answer message (step 224), and the communication terminal 100 reproduces and displays the answer at least once (step 226). The program may be configured to display. As a result, the burden on the operator is reduced, and the efficiency of work can be improved.

  Alternatively, the program may be configured to display a telephone icon after the dialog server determines that the answer message in the dialog server cannot respond to the user's question even if the user does not switch to the call mode. . In this case, the dialog server notifies the user that the call mode is to be switched (step 400), and at the same time, accesses the communication service cooperation server to connect the telephone line between the user and the operator or the operator (step). 402).

  Alternatively, as a result of analyzing the user's question, a telephone icon may be displayed on the display unit 106 after the dialog server determines that a dialog between the user and the operator is essential, and the user may be prompted to switch to the call mode. .

  When the mode is shifted to the call mode, the contents and history (log) of dialogue between the user and the server may be stored and displayed to the operator in charge (step 406). Thereby, the operator can consider the answer to the user's question more efficiently. In addition, it is less necessary for the user to repeat the conversation that has already been performed. For this reason, the business operator can promptly provide the user with a higher quality and more profound service.

  When the communication service cooperation server connects the telephone from the user to the communication network, the contents of the dialogue are analyzed (step 408), the most appropriate operator is selected based on the result (step 410), and the user and the selected operator are selected. A telephone communication may be established between and (step 412). This allows the user to talk to the most appropriate operator for the question.

[5. Transition to map mode]
As described above, the program may be configured to display the icon 302 for accessing the operator's store on the display unit 106 of the user's communication terminal 100 (see FIG. 3). When the user operates this icon 302, it is possible to shift to the map mode. A specific flow is shown in FIG. Each step shown in FIG. 5 is an example, and for example, the order of these steps may be appropriately changed.

  When the map mode is activated by the user's operation (step 500) (step 502), the communication terminal 100 takes in the position coordinates using a built-in GPS (global positioning system) function according to the instructions of the program. Alternatively, the user himself / herself inputs an address or nearby landmark information into the voice input unit 110 as voice information or by using the input unit 108 as character information, thereby capturing the position coordinates of the communication terminal 100. The program may be configured as described above. Thereby, the position of the communication terminal is specified (step 504). When the communication terminal 100 is not equipped with a GPS function or in an environment where the GPS function cannot be used, the communication terminal 100 communicates with a base station of a nearby communication service provider to locate the communication terminal 100 The program may be configured so that the identification is performed. In this case, the position of the communication terminal 100 is specified using a position information specifying server, a CTI (Computer Telephony Integration) server, or the like. The location specifying server and the CTI server may be constructed in a server managed and provided by an information processing service provider or a communication service provider.

  After the position of the communication terminal 100 is specified, the communication terminal 100 accesses the information distribution server operated by the provider providing the map information and acquires the map data according to the instructions of the program, and displays it on the display unit 106 (Step 506). The program further causes the communication terminal 100 to acquire the location information of the store stored in the server and the service content provided by the store (step 508), and displays the store on the display unit 106 based on this information (step 508). 510). For example, as shown in FIG. 6, a map near the communication terminal 100 is displayed on the display unit 106, and icons 602 and 604 representing stores are displayed.

  Not only stationery, stationery information such as specific service contents, functions, and business hours provided by the store, but also the congestion status and the amount of resources (product inventory, amount of human resources that can be allocated to users, etc.) The program may be configured so that the business operator sequentially stores parameters that change sequentially with time in a database in the server, and the communication terminal 100 acquires and displays these information and parameters on the display unit 106 ( Steps 508, 510). In this case, you may make it display the said information and parameter regarding each store by operating the icon of a store. As a result, the user can easily grasp the amount of resources managed by the business operator and the variation thereof.

  Furthermore, in the information providing method of the present embodiment, a route from the current position of the user to the store that the user desires to access may be displayed. For example, the communication terminal 100 transmits the position coordinates of the user and the position information of the target store to the information distribution server, requests a search for an appropriate route (step 512), and receives the route searched in the information distribution server. Then, the route may be displayed on the display unit 106 (step 514). For example, as shown in FIG. 7, the route can be displayed by using icons to indicate the current location of the user and the route to the target store. Here, the route is shown using a plurality of circular icons. In addition, the distance to the destination and the required time may be displayed.

  Here, an example in which the user switches to the map mode by operating the icon 302 (step 500) has been described, but the switch to the map mode may be performed based on the dialogue between the user and the virtual operator. For example, if the dialog server analyzes the dialog between the user and the virtual operator and determines that the final answer to the user's question includes a guide to the store (step 516), the dialog server can respond appropriately to the user's question. An appropriate store is selected, a guide to the store is notified (step 518), the map mode is activated (step 502), the current location of the user and the location information of the corresponding store on the display unit 106, and the route to the store Etc. may be displayed.

  As described above, according to the information providing method of the present embodiment, it is possible to narrow down the stores suitable for the service requested by the user through the dialogue between the user and the virtual operator, and to provide a high-quality service to the user. That is, instead of providing a plurality of answers to the user's question and allowing the user to select, the user can be guided to an answer suitable for the user through the dialogue between the user and the virtual operator.

[6. Transition to web chat mode]
A mode that provides dialogue with text input and output without providing voice input or output depending on the user's surrounding environment, communication terminal usage conditions, or the user's own preferences, and providing information requested by the user through this (referred to below) It may be possible to shift to a web chat mode. For example, when the user himself / herself is prohibited from uttering, when voice transmission from the communication terminal is prohibited, when the user's voice and ambient noise cannot be distinguished due to a busy environment, the user can hear the voice of the communication terminal. When the output unit 112 is used for other purposes (such as music or video viewing), the program may be configured so that the user can automatically shift to the web chat mode. In the web chat mode, both the question entered by the user and the answer message are displayed on the display unit 106 in time order. An example of the flow at this time is shown in FIG.

  For example, after the communication terminal 100 is activated (step 200), an icon for shifting to the web chat mode is displayed on the screen (see FIG. 3) on which the virtual operator is displayed (step 700) so that the user can arbitrarily operate it. A program may be configured. When the user operates this icon (step 702), the screen on which the virtual operator is displayed is switched to a screen (web chat screen) on which interactive text is displayed (step 704). . Thereafter, the user can start the dialogue by inputting text (step 706).

  In this case, the communication terminal 100 that does not send the voice information from the voice synthesis server or receives the voice information does not reproduce the voice information, and only displays the text of the voice information on the display unit 106. Alternatively, the reply message selected or generated by the dialog server may be transmitted to the communication terminal 100 as the character information.

  Alternatively, after the application is started (step 200), the program instructs to check the usage status of the audio output unit 112 (step 710), and if it is determined that the audio output unit 112 is used for other purposes, the web Whether or not to switch to the chat mode is displayed on the display unit 106 (step 712), and the mode shift by the user may be selected.

  Alternatively, the program may be configured to switch to the web chat mode according to a change in some physical quantity, for example, an environmental parameter around the communication terminal 100. For example, the sound volume around the communication terminal 100 is measured, and the sound volume data is transmitted to a dialogue server or the like (step 720), and if it is above a certain level, the display unit 106 displays that the mode is shifted to the web chat mode (step 722). ), The mode may be automatically shifted to the web chat mode (step 704). The surrounding sound volume can be measured, for example, after the application is started and before the user activates the voice input unit 110. The threshold value for shifting to the web chat mode may be set to 90 dB or more, or 110 dB or more, for example. Further, the determination of the transition to the web chat mode may be performed on the communication terminal 100 side or the conversation server side.

  Although not shown, when the noise around the communication terminal 100 and the voice information input by the user cannot be distinguished, the program may be configured to shift to the web chat mode. For example, when the voice recognition server analyzes the voice information and detects the voice section, the section including the user's voice information cannot be detected, or the detected section does not reach a certain time, or the input voice When noise cannot be sufficiently removed from information and a user's voice (question) cannot be detected and analyzed, an icon for shifting to the web chat mode is displayed on the display unit 106 at the same time as guiding the shift to the web chat mode. Also good. Alternatively, an answer message requesting the user to move to an environment with less noise may be transmitted to the communication terminal 100 and an icon may be displayed.

Moreover, you may comprise a program so that it may transfer to web chat mode based on the content of interaction. For example, if the conversation server determines that the user's voice information contains sensitive personal information, or if the conversation server needs to include such personal information in an answer message that it selects and generates, The transition to the web chat mode may be guided.
[7. Transition to other modes]
The mode shift is not limited to the above-described call mode, map mode, and web chat mode. For example, based on analysis of interaction between the user and the virtual operator, or based on the user's judgment, the program can access various services (hereinafter referred to as network service) provided by the operator via the communication network. It may be configured. When allowing access to the network service at the user's own discretion, for example, an appropriate icon is displayed on the display shown in FIG. 3, and the user can actively access the network service by operating the icon. Also good. Examples of the network service include an internet banking service provided by each bank in the case of the financial industry.

  On the contrary, when the optimum answer selected by the dialog server is the use of the network service by the user, an answer message from the dialog server is transmitted to the communication terminal 100 and an icon linked to the network service is displayed on the display unit 106. May be.

  As described above, in the information providing method of the present embodiment, when a certain condition is satisfied, or when some physical quantity, for example, a noise level or a noise frequency around the communication terminal 100 changes, or a user desires In this case, it is possible to shift to various modes. The mode transition can be performed voluntarily by the user or automatically. Alternatively, it is possible to prompt the user for the mode through dialogue. As described above, the user interface is switched from the uppermost layer (application layer) where the virtual operator is displayed at the time of activation to various lower layers, and information and answer messages suitable for the user's question can be provided.

Second Embodiment
In the present embodiment, a system configuration for realizing the information providing method described in the first embodiment will be described with reference to the drawings. Since the system configuration of the communication terminal 100 has been described in the first embodiment (see FIG. 1), it is omitted here.

  The system configuration described in the present embodiment is an example, and the present invention is not limited to the embodiment described below. Functions and elements of various servers described below may be duplicated with other servers. For example, a database may be managed and used by a plurality of servers. A unit (such as a controller) that controls each server may be shared by a plurality of servers and configured to control the plurality of servers. Furthermore, one server may be a component of another server. In this case, the one server can be recognized as a part of another server. For example, when the server A having a certain function is one of the components of the server B, the server A can be recognized and described as one functional unit in the system including the server B.

[1. voice recognition]
The voice input by the user is captured by the voice input unit 110 of the communication terminal 100 and transmitted to the communication unit 114 as voice information (see FIG. 1). The communication unit 114 transmits voice information to the API 810 serving as an interface of the voice recognition server 800 illustrated in FIG. 9 via the communication network. The transmitted voice information is analyzed by the voice recognition engine 820 in the voice recognition server 800.

  As illustrated in FIG. 9, the speech recognition engine 820 includes, for example, a feature amount extraction unit 830, a decoder 840, and a text output unit 850. The feature quantity extraction unit 830 analyzes the transmitted voice information and extracts feature quantity parameters. The decoder 840 calculates the best candidate from the feature parameter while referring to databases such as the language model 842, the acoustic model 844, the vocabulary dictionary 846, and outputs the result as a recognition result. The output recognition result is converted into text data by the text output unit 850 and output.

  By using such a speech recognition system, a question spoken by the user in spoken language (spoken language) to the virtual operator is converted into text.

[2. Speech synthesis]
Voice reproduced by the user's communication terminal 100 is transmitted from the API 910 serving as an interface of the voice synthesis server 900 to the communication terminal 100 via the communication network (see FIG. 10). The speech synthesis server 900 includes a speech synthesis engine 920. As illustrated in FIG. 10, the speech synthesis engine 920 includes, for example, a text input unit 930, a language analysis unit 940, a waveform generation unit 950, and a transmission unit 960. ing.

  In the text input unit 930, character information of a message selected by the dialogue server described later is input. This character information is subjected to language analysis in the language analysis unit 940, and the accent position and reading information are analyzed while referring to the language dictionary 942 which is a database. The waveform generation unit 950 refers to a speech synthesis database 944 created using waveform data obtained from actually recorded speech, and generates speech waveforms. The voice data generated by the waveform generation unit 950 is transmitted by the transmission unit 960 to the communication terminal 100 via the AIP 910. Note that the transmission unit 960 can transmit not only voice data but also input character information as characters (text).

  By using such a speech synthesis system, the virtual operator can interact with the user in the same voice as a person speaks, reducing the sense of distance to the operator by making the user feel close Can be made.

[3. Dialogue]
After the voice message transmitted from the user is analyzed by the voice recognition server 800 and converted into character information, an appropriate answer is generated and selected by the dialogue server 1000 (see FIG. 11). In addition to the API 1010, the dialogue server 1000 includes a dialogue engine 1020, a provider management server 1040, a management server 1050, an information database 1060, and the like (see FIG. 11).

  The dialogue engine 1020 can include a dialogue controller 1022, a selector 1024, a response generation unit 1026, and the like. The dialogue controller 1022 controls the dialogue engine 1020, and has a function of grasping the user's intention and causing the selector 1024 to select an appropriate message. The selector 1024 has a function of narrowing down and selecting an appropriate answer message or inquiry message (1032, 1034, 1036, 1038) for the user's intention understood by the dialog controller 1022. The selector 1024 is also called a refinement engine. The response generation unit 1026 has a function of creating an answer to the user based on the selected message. The response generation unit 1026 may create a single message by fusing a plurality of messages.

  The information database 1060 can store terms and knowledge related to the business of the operator and is referred to by the dialogue engine 1020. The business management server 1040 can include a database 1042 that can store parameters such as store functions and location information, contents of services that can be provided, and parameters that vary with time, such as store congestion and product inventory. . The database 1042 may store personal information such as standard data of user biometric information. The response generation unit 1026 may generate an answer message by appropriately merging personal information stored in the database 1042 with the selected response message. In the case where user information is stored in the database 1042, the operator management server is preferably managed by the operator.

  The dialogue server may be connected to a communication service cooperation server 1100 (described later) that stores dialogue contents and history between the user and the dialogue server. Alternatively, the system may be configured to store the above-described conversation contents and history in the database 1042.

  Messages (1032, 1034, 1036, 1038) are answer messages or inquiry messages corresponding to various questions from the user, and can be prepared and managed as character information. For example, the message 1032 can be a standard reply message corresponding to FAQ. The message 1034 can be, for example, an inquiry message for asking a question when the answer message cannot be narrowed down. The message 1036 can be a message for responding to a chat between the user and the operator. The messages 1032, 1034, and 1036 can be edited on the operator management server 1040. The message 1038 can be edited by the management server 1050 managed by the operator that manages and maintains the voice recognition server 800 and the dialogue server 1000, and is for a dialogue that is not directly related to the business of the user's operator. To be prepared.

  By preparing the inquiry message 1034, the user can be guided to an answer desired by the user without imposing a burden on the user to select from many answers.

  By preparing and editing a message 1036 for responding to a chat, the user can perform various conversations with the virtual operator even if the conversation is less relevant to the business content of the business. This makes it possible to communicate more closely with the operator.

[4. Call mode]
When shifting from the dialogue between the user and the virtual operator to the call mode, the communication terminal 100 is connected to the communication service cooperation server 1100 via the API 1110 which is an interface (see FIG. 12). The communication service cooperation server 1100 includes a communication status management database 1130, a window database 1140, and the like.

  The communication status management database 1130 can store a history of dialogue and contents of dialogue between the user and the virtual operator, and this content can be browsed by the operator or the operator on the operator management server 1040. The window database 1140 can store information related to the operator, for example, information such as a department to which the operator belongs, a business scope, knowledge and experience. The controller 1120 analyzes the dialogue log and dialogue contents between the user and the virtual operator, and based on the result, the controller 1120 can refer to the window database 1140 and select the optimum operator. Note that the communication status management database 1130 and the window database 1140 may not be installed in the communication service cooperation server 1100, but the information stored in these servers may be managed by the operator management server 1040.

  When shifting to the call mode, a request for telephone connection is transmitted to the communication network 1150 of the provider providing the telephone communication service via the API 1110, and at the same time, the telephone number of the connection destination is transmitted. As a result, a call between the user and the operator or the operator becomes possible.

[5. Map mode]
As shown in FIG. 13, the system for shifting to the map mode can be configured by an information distribution server 1200, a dialogue server 1000, an operator management server 1040, etc. operated by an operator providing map information, for example. .

  The information distribution server 1200 includes a map information database 1230. The communication terminal 100 can transmit position information to the information distribution server 1200 and receive map information from the information distribution server 1200 via the API 1210. The business management server 1040 can include a database 1042 that can store information such as store functions and location information, the contents of services that can be provided, and parameters that vary with time, such as congestion status and product inventory status. The dialogue server 1000 can analyze a question from the user and select an optimal store while referring to the database 1042 based on the result. The communication terminal 100 is connected to the dialog server 1000 via the API 1010, and acquires the store selected by the dialog server 1000, information related thereto, and parameters. The map information received from the information distribution server 1200 and the location information of the selected store are displayed on the display unit 106 in accordance with the program instructions.

  The information distribution server 1200 includes route search means 1220. The route search means 1220 includes, for example, a route search means 1220, a search database 1240, and a road traffic information database 1250 in addition to the map information database 1230. The route search means 1220 has a function of searching for and determining the best route to the user and the store while referring to information stored in the map information database 1230, the search database 1240, and the road traffic information database 1250.

  In the present embodiment, a bank is taken as an example of a business operator, and a case where a user exchanges Japanese yen into US dollars will be described with reference to FIG.

  First, the user starts an application including the program shown in the first embodiment installed in the communication terminal 100. Thereafter, the voice input unit 110 is activated by operating the icon 300 (see FIG. 3), and the question is verbally described in a manner of interacting with the virtual operator. For example, “I want to exchange Japanese yen”. If it is difficult to speak, a question may be input by text input. The communication unit 114 of the communication terminal 100 transmits the user's voice information to the voice recognition system according to the program command (S1). The voice recognition server 800 converts the voice information into character information and transmits it to the dialogue server 1000 (S2).

  The dialogue server 1000 uses the dialogue engine 1020 to understand the intention of the question. However, since the exemplified voice information does not include information such as the type of foreign currency and the amount of exchange, the business operator provides the best information to the user. Can not provide. In order to obtain further information from the user, the dialogue server selects or newly creates an optimal inquiry message from the prepared inquiry messages, and transmits it to the speech synthesis server 900 (S3). For example, an inquiry message such as “Tell me the type of foreign currency” or “Tell me the amount of Japanese yen to exchange” is selected or created.

  The speech synthesis server 900 performs a linguistic analysis on the inquiry message to generate a waveform, and transmits the waveform as speech information to the communication terminal 100 (S4). Note that the character information of the inquiry message may be transmitted from the speech synthesis server 900 or the dialogue server 1000 to the communication terminal 100 simultaneously with the speech information. The communication terminal 100 reproduces voice information using the voice output unit 112 and displays an inquiry message on the display unit 106.

  The user responds to the inquiry message in the same manner as in step S1. Possible answers include, for example, “I want to exchange for US dollars” and “Japanese yen is about 100,000 yen”. The communication terminal 100 transmits this answer to the voice recognition server 800 (S5), and the voice recognition server 800 converts the user's voice into character information and transmits it to the dialogue server 1000 (S6).

  The dialogue server 1000 analyzes the answer from the user. In this embodiment, since the exchange is from Japanese yen to US dollars, the user needs to access the store directly. For this reason, the dialogue server 1000 determines that it is essential to guide the user to the store, and generates or selects an answer message for guiding the user to the store. An example of the reply message is a message such as “I will guide you to a nearby branch office”. Thereafter, the voice information of the reply message is transmitted to the voice synthesis server 900 (S7).

  At this time, the dialogue server 1000 accesses the operator management server 1040 (S8), and acquires store candidates that satisfy the user's request, and parameters such as store congestion and product inventory (S9). This is because the type and amount of foreign currency to be stored differs depending on the store, and the inventory amount may fluctuate greatly depending on the transaction status. Based on these pieces of information, an optimal store for the user's request is determined, and the position information is transmitted to the communication terminal 100. At the same time, the dialogue server 1000 transmits a request for shifting to the map mode to the communication terminal 100 (S10).

  The voice synthesis server 900 generates voice information from the reply message, and transmits the voice information and character information of the reply message to the communication terminal 100 (S11).

  The communication terminal 100 acquires position coordinates using the built-in GPS function, transmits the position information of the communication terminal 100 and the received position information of the store to the information distribution server 1200, and requests a route search together with nearby map information. (S12). In response to the request from the communication terminal 100, the information distribution server 1200 transmits necessary map information and a route search result to the communication terminal 100 (S13).

  Thereafter, the communication terminal 100 displays a map of the vicinity, the position of the target store, and the route to the store on the display unit 106 according to the instructions of the program.

  Thus, by responding to the user's question through dialogue, an optimal answer can be provided to the user. At this time, the user does not need to pay the effort of selecting the optimum answer from a plurality of answers, and can quickly reach the optimum answer.

  In addition, since the service based on such an interaction can be achieved by using the communication terminal 100 operated by the user, the user does not have to pay labor to visit the store and ask a question. There is no need to consider time. Thus, by using the information providing method of the present embodiment and the program for executing it, the user can solve various problems regardless of time and place.

  In addition, the user can acquire uniform and abundant information without being restricted by the knowledge, experience, and ability of the business operator and its operator. For example, in the case of this example, the US dollar cash inventory status at each store changes from moment to moment, so it has been difficult to grasp the fluctuations by operators who directly correspond to the user, and the inventory is insufficient. There was a possibility to guide a special store. However, by implementing the information providing method of this embodiment, accurate information can be provided to all users. At the same time, business operators can achieve more efficient operations and improved productivity by avoiding the consumption of human resources in response to users.

  In the present embodiment, a description will be given assuming a case different from the first embodiment. Specifically, an example in which the user has lost a bank account cash card will be described with reference to FIG.

  Similar to the first embodiment, the user first starts an application including the program shown in the first embodiment installed in the communication terminal 100. Thereafter, the voice input unit 110 is activated by operating the icon 300, and the question is verbally described in a manner of interacting with the virtual operator. For example, you can say “I lost my cash card”. If it is difficult to speak, a question may be input by text input. According to the command of the program, the communication unit 114 of the communication terminal 100 transmits the user's voice information to the voice recognition system (FIG. 15, S1). The voice recognition server 800 converts the voice information into character information and transmits it to the dialogue server 1000 (S2).

  The dialogue server 1000 uses the dialogue engine 1020 to understand the intention of the question. The dialogue server 1000 selects an answer message for the question, and transmits the character information to the speech synthesis server 900 (S3). The speech synthesis server 900 performs a linguistic analysis on the inquiry message to generate a waveform, and transmits the waveform to the communication terminal 100 as speech information and character information (S4). The communication terminal 100 reproduces voice information using the voice output unit 112 and displays an inquiry message on the display unit 106.

  In the case assumed in this embodiment, the business operator needs to promptly notify the user to a predetermined department, for example, the loss reception center. Therefore, the dialog server selects the reply message and requests the business management server 1040 to transmit the telephone number of the corresponding department (S5). The business entity management server 1040 transmits necessary information in accordance with a request from the dialogue server 1000 (S6).

  The reply message in the case of the present embodiment is, for example, a message such as “I will make a transaction with a cash card, so I will connect to the loss reception center”. Thereby, the user can predict switching to the call mode.

  The dialogue server 1000 requests the communication terminal 100 to switch to the call mode (S7). At the same time, the dialog server 1000 requests the communication service cooperation server 1100 to connect the line and notifies the telephone number of the corresponding department (S8). In response to a request from the dialogue server 1000, the communication service cooperation server 1100 connects the user's communication terminal 100 and a predetermined department via a telephone line (S9), and a call between the user and the operator is started.

  In addition, although the example which switches to telephone call mode was shown here irrespective of a user's intention, you may entrust the necessity of the switching to telephone call mode to a user's judgment. For example, when the communication terminal 100 receives a request for switching to the call mode from the dialogue server 1000 (S7), a telephone icon may be displayed on the display unit 106, and the telephone call may be started by operating the telephone icon.

  The question assumed in this embodiment is a kind of FAQ, and for example, it is often possible to obtain an answer even on a site provided by a business operator. However, the case of the present embodiment is a kind of emergency situation for the user, and since it is desired to solve the problem as soon as possible, it often does not have a mental margin for accessing the site of the operator from the communication terminal 100.

  On the other hand, in the present embodiment, it is possible to automatically start a call to a predetermined department simply by starting an application installed in a communication terminal 100 that is always portable, such as a smartphone, and simply speaking a question. It is. Therefore, according to the embodiment of the present invention, it is possible to provide a simpler and quicker service to the user.

  In the above-described embodiments and examples, description has been given mainly taking examples related to the financial business, but the implementation of the present invention is not limited to the financial industry. For example, it can be implemented and applied in all industries and industries such as the food industry, restaurant industry, logistics industry, construction industry, airport and shipping industry. As an example, in the present embodiment, a case where a user uses a restaurant is assumed, and description will be made with reference to FIG. Here, the operator is an individual or an organization that collects information about restaurants and provides them to users, and each restaurant provides necessary information to the operators.

  As in the first and second embodiments, the user first starts an application including the program shown in the first embodiment installed in the communication terminal 100. Thereafter, the voice input unit 110 is activated by operating the icon 300, and the question is verbally described in a manner of interacting with the virtual operator. For example, you can say "I'm hungry" or "I want to eat rice". If it is difficult to speak, a question may be input by text input. The communication unit 114 of the communication terminal 100 transmits the user's voice information to the voice recognition system according to the program command (S1). The voice recognition server 800 converts the voice information into character information and transmits it to the dialogue server 1000 (S2).

  Although the dialogue server 1000 uses the dialogue engine 1020 to understand the intention of the question, the above voice information does not contain any specific information. For example, it does not contain any information such as what the user wants to eat, how much he wants to eat, whether he wants to drink alcohol, how many people are, whether he is a smoker, etc. Therefore, the business operator cannot provide the best information to the user. In order to obtain further information from the user, the dialogue server selects an optimum message from the prepared inquiry messages or creates a new message and transmits it to the speech synthesis server 900 (S3). The inquiry message may be a free-answer message such as “What do you want to eat?” Or a choice-type message such as “Which is Japanese, Chinese or Western?” .

  The speech synthesis server 900 performs a linguistic analysis on the inquiry message to generate a waveform, and transmits the waveform as speech information to the communication terminal 100 (S4). Similarly to the first and second embodiments, text information may be transmitted from the speech synthesis server 900 or the dialogue server 1000 to the communication terminal 100 simultaneously with the speech information. The communication terminal 100 reproduces voice information using the voice output unit 112 and displays an inquiry message on the display unit 106.

  At the same time, the dialogue server 1000 requests the communication terminal 100 to provide location information (S5). In response to this request, the communication terminal 100 acquires position coordinates using the built-in GPS function, and transmits the information to the dialogue server 1000 (S6).

  During this time, the user answers the inquiry message in the same manner as in step S1. The answer may be a concrete answer such as “I want to eat pasta” or an abstract answer such as “Warm food is good”. The communication terminal 100 transmits this answer to the voice recognition server 800 (S7), and the voice recognition server 800 converts the user's voice into character information and transmits it to the dialogue server 1000 (S8).

  The dialogue server 1000 analyzes the answer from the user. When the dialog server 1000 determines that it is difficult to narrow down to the final restaurant based on the answer from the user, it selects and creates an inquiry message again and transmits it to the communication terminal 100 via the speech synthesis server 900. And the user verbally gives an answer to that. By repeating this step, for example, “I want to eat warm food”, “Good Japanese food”, “A place to serve sake”, “3 people”, “A place where you can smoke” Specific information can be acquired from the user.

  The dialogue server 1000 requests the operator management server 1040 for information on restaurants registered near the position coordinates of the user (S9). In response to this request, the business entity management server 1040 transmits the information requested from the database 1042 to the dialogue server 1000 (S10).

  The dialogue server 1000 analyzes the information obtained from the user, and selects an answer that best fits the user's wishes from the information obtained from the database 1042 using the narrowing function of the selector 1024.

  Next, the dialogue server 1000 selects an answer message for informing the user that an answer has been obtained, and transmits this to the speech synthesis server 900 (S11). As an answer message at this time, for example, a message such as “I will introduce the best shop for customer's request” may be used. At the same time, the dialogue server 1000 requests the information distribution server 1200 for the selected answer, that is, the location information of the selected restaurant (S12). The information distribution server 1200 transmits the requested information from the dialogue server 1000 to the dialogue server 1000 or directly to the communication terminal 100 (S13).

  The voice synthesis server 900 converts the reply message sent from the dialogue server 1000 into voice information and sends it to the communication terminal 100 (S14).

  Furthermore, the dialogue server 1000 requests the communication terminal 100 to make an intention for the map mode (S15). In response to this, the communication terminal 100 requests the information distribution server 1200 for a route search together with the position information of the communication terminal 100 and the position information of the selected restaurant (S16). In response to the request from the communication terminal 100, the information distribution server 1200 transmits necessary map information and a route search result to the communication terminal 100 (S17).

  Thereafter, the communication terminal 100 displays a map of the vicinity, the position of the target restaurant, and the route to the restaurant on the display unit 106.

  In this way, by responding to the user's question through dialogue, an optimal answer can be provided to the user. At this time, the user does not need to pay the effort of selecting the optimum answer from a plurality of answers, and can quickly reach the optimum answer.

  The above-described embodiments can be implemented in appropriate combination as long as they do not contradict each other. In addition, components that are appropriately added, deleted, or changed in design by those skilled in the art based on each embodiment are also included in the scope of the present invention as long as they include the gist of the present invention.

  In addition, even for other operational effects different from the operational effects brought about by the aspects of the above-described embodiments, those that are apparent from the description of the present specification, or that can be easily predicted by those skilled in the art, Of course, it is understood that the present invention provides.

  100: Communication terminal, 102: Central processing unit, 104: Memory, 106: Display unit, 108: Input unit, 110: Audio input unit, 112: Audio output unit, 114: Communication unit, 116: Image capturing device, 200 : Step, 202: Step, 204: Step, 206: Step, 208: Step, 210: Step, 212: Step, 214: Step, 216: Step, 218: Step, 220: Step, 222: Step, 224: Step 226: Step, 228: Step, 230: Step, 232: Step, 234: Step, 300: Icon, 302: Icon, 400: Step, 402: Step, 404: Step, 406: Step, 408: Step, 410 : Step, 412: Step, 602: A 604: Icon, 700: Step, 702: Step, 704: Step, 706: Step, 710: Step, 712: Step, 720: Step, 722: Step, 800: Voice recognition server, 810: API, 820: Speech recognition engine, 830: feature extraction unit, 840: decoder, 842: language model, 844: acoustic model, 846: vocabulary dictionary, 850: text output unit, 900: speech synthesis server, 910: API, 920: speech synthesis Engine, 930: text input unit, 940: language analysis unit, 942: language dictionary, 944: speech synthesis database, 950: waveform generation unit, 960: transmission unit, 1000: dialog server, 1010: API, 1020: dialog engine, 1022: Dialog controller, 1024: Selector, 1 26: Response generation unit, 1032: Message, 1034: Message, 1036: Message, 1038: Message, 1040: Operator management server, 1042: Database, 1050: Management server, 1060: Information database, 1100: Communication service cooperation server, 1110: API, 1120: Controller, 1130: Communication status management database, 1140: Window database, 1150: Communication network, 1200: Information distribution server, 1210: API, 1220: Route search means, 1230: Map information database, 1240: Search Database 1250: Road traffic information database

Claims (33)

A voice recognition unit for recognizing voice information of a user transmitted from a communication terminal;
A dialog unit for selecting a reply message to the voice information from a database in which a plurality of messages are stored;
A transmission unit that transmits the selected response message to the communication terminal as at least one of voice information and character information;
The information providing system in which the dialog unit selects an inquiry message for asking the user a question from the database when the answer message cannot be narrowed down to one, and transmits the selected inquiry message to the communication terminal.   The information providing system according to claim 1, further comprising the communication terminal. The communication terminal is configured to display a virtual operator;
2. When the response message and the inquiry message are reproduced on the communication terminal, a part of the virtual operator is configured to perform a large operation compared to when the answer message and the inquiry message are not reproduced. 1. The information providing system according to 1. A first server storing the standard biometric information of the user;
The communication terminal has an image capturing device that captures the user's biological information,
The information providing system according to claim 1, wherein the user is allowed to input the audio information when the user is approved based on the biometric information and the standard biometric information.   The information providing system according to claim 1, further comprising a second server for a call between the user and a business operator.   The information providing system according to claim 5, wherein the second server is configured to store the voice information recognized by a voice recognition unit, the selected answer message, and the inquiry message. A third server storing map information;
The first server is configured to store information regarding a plurality of stores of the operator;
The information providing system according to claim 4, wherein the communication terminal is configured to display a position of the communication terminal and information of the plurality of stores on a display unit. The dialogue unit is configured to select at least one store from the plurality of stores based on the audio information,
The information providing system according to claim 7, wherein the third server is configured to transmit a route to the selected store to the communication terminal.   If the answer message corresponding to the question inputted by voice by the user is narrowed down from a plurality of messages, and the answer message cannot be narrowed down to one, the inquiry message for asking the user a question is selected from the plurality of messages. Server configured to.   The server according to claim 9, wherein the question is transmitted from a voice recognition server that recognizes voice information input by a user.   The server of claim 9, wherein the answer message and the inquiry message are configured to be communicated to a speech synthesis server.   The server according to claim 9, connected to a first server configured to store biometric information of the user.   The server of claim 12, wherein the first server is configured to store business store information.   14. A second server configured to store at least one of the user's question, the selected answer message, and the selected inquiry message is connected to the first server. Server described in. Recognize voice information including user questions sent from communication terminals,
Selecting a response message corresponding to the question of the recognized user from a plurality of response messages;
The selected response message is transmitted to the communication terminal as at least one of voice information and character information,
If the answer message cannot be narrowed down to one, select an inquiry message for asking a question from the plurality of answer messages,
A program for causing a computer to execute transmission of the selected inquiry message to the communication terminal.   The program according to claim 15, wherein the user's biometric information captured by the communication terminal is received and the user is authenticated using the biometric information.   The program according to claim 15, further comprising storing at least one of the question of the user, the selected answer message, and the selected inquiry message. Refer to at least one of the stored question of the user, the selected answer message, the selected inquiry message;
The program according to claim 15, further comprising connecting the communication terminal and a telephone terminal of a business operator. Refer to at least one of the stored question of the user, the selected answer message, the selected inquiry message;
The program according to claim 15, further causing the store information of the business operator to be displayed on the communication terminal. Receiving the question as text information from the communication terminal;
The program according to claim 15, further causing the reply message to be transmitted as character information to the communication terminal. When the question included in the voice information cannot be detected, the question is received as character information from the communication terminal,
The program according to claim 15, further causing the reply message to be transmitted as character information to the communication terminal. Receiving volume data around the communication terminal from the communication terminal;
The program according to claim 21, wherein the program is executed to receive the question as character information from the communication terminal according to the volume data. Sends speech information to a system that includes a speech recognition server, dialogue server, speech synthesis server,
Receiving at least one of an answer message to the user's question included in the audio information, an inquiry message to the question,
A program for causing a communication terminal to output the answer message and the inquiry message as at least one of voice information and character information.   24. The program according to claim 23, wherein the program is configured to be stored in the communication terminal.   The program of claim 23, wherein the program is configured to be obtained from a communication network.   When the program displays a virtual operator and the answer message and the inquiry message are reproduced on the communication terminal, a part of the virtual operator performs a large operation compared to when the answer message and the inquiry message are not reproduced. 24. The program of claim 23, configured to perform.   24. The program of claim 23, wherein the program is configured to obtain user biometric information.   The program according to claim 23, wherein the program is configured to connect the communication terminal and a telephone terminal of a business operator. The program acquires location information of the communication terminal,
Displaying a map on the communication terminal;
The program according to claim 23, configured to display store information of a business operator on the communication terminal.   30. The program according to claim 29, wherein the program further displays a route from the position of the communication terminal to the business operator's store.   The program according to claim 23, further causing the question to be transmitted as character information.   24. The program according to claim 23, wherein a volume around the communication terminal is measured, and the volume data is transmitted to at least one of the voice recognition server, the dialogue server, and the voice synthesis server.   The program according to claim 32, further causing the question to be transmitted as character information in accordance with the volume.