KR102136706B1 - Information processing system, reception server, information processing method and program - Google Patents

Abstract

The object of the present invention is to make the task of accepting inquiries from customers more efficient. In the information processing system 1, the speech recognition unit 104 accepts an inquiry from the terminal device 70. The intention/status analysis unit 107 determines whether or not an answer to the inquiry can be searched from the FAQ search database 154, and when it is determined that the answer cannot be searched, information for performing the answer. A replenishment process to replenish is executed, and an answer is generated based on the result of the replenishment process. The intention/status analysis unit 107 determines whether or not an answer to the inquiry has been generated, and if it is determined that an answer can be searched for and if it is determined that an answer to the inquiry has been generated, it is searched for. Send a reply or a generated reply. The communicator call unit 112 executes the call processing of the communicator when it is determined by the intention/state analysis unit 107 that an answer cannot be generated.

Description

Information processing system, reception server, information processing method and program

The present invention relates to an information processing system, a reception server, an information processing method and a program.

Conventionally, a technique for automating a call center task for accepting inquiries from customers is known. For example, a system has been known that responds to inquiries by telephone from a customer to a call center by automatic voice response (IVR). In addition, a system is known in which a virtual operator (VO) accepts inquiries from customers from the call center to the Internet or the like. In addition, the technique of automating a call center task is described in patent document 1, for example.

Patent Document 1: Japanese Patent Publication No. 2014-178381

However, in the conventional technique of automating the call center business, the usage form is limited, such as the automated business is limited to the distribution of Q&As. Therefore, the task of accepting inquiries from customers has not been sufficiently efficient.

The present invention has been made in light of this situation, and aims to make the task of accepting inquiries from customers more efficient.

In order to achieve the above object, an information processing system of one aspect of the present invention,

An information processing system comprising a terminal device for sending an inquiry and a reception server for accepting the inquiry sent from the terminal device,

The reception server,

Reception means for accepting the inquiry from the terminal device;

First determination means for determining whether an answer to the inquiry can be retrieved from a response database in which a plurality of questions and answers prepared in advance are stored;

When it is determined by the first determination means that the answer to the inquiry cannot be retrieved, a supplementary process is performed to supplement the information for performing the answer to the inquiry, and based on the result of the supplementary process, Answer generating means for generating a reply to the inquiry;

Second determination means for determining whether or not an answer to the inquiry could be generated by the answer generation means;

When it is determined by the first determination means that an answer to the inquiry can be retrieved, and when it is determined by the second determination means that an answer to the inquiry can be generated, by the first determination means Reply means for transmitting the searched reply or the reply generated by the second determination means;

When it is determined by the second determination means that an answer to the inquiry cannot be generated, a calling means for executing a call processing of a communicator corresponding to the inquiry

It characterized in that it comprises a.

According to the present invention, the task of accepting inquiries from customers can be made more efficient.

1 is a diagram showing a system configuration of an information processing system according to the present invention.
2 is a diagram showing a hardware configuration of a server computer constituting each server.
3 is a diagram showing a hardware configuration of a terminal device.
4 is a block diagram showing a main functional structure realized in an information processing system.
5 is a flowchart showing the flow of the acceptance processing executed by the information processing system.
6 is a diagram showing an example of a display screen displayed on the terminal device in the reception processing.
7 is a flowchart showing the flow of VOC collection processing.
Fig. 8 is a flowchart showing the flow of the animation management process executed by the information processing system.
9 is a block diagram showing a main functional structure realized in the information processing system according to the second embodiment.
10 is a flowchart showing the flow of the cooperative processing executed by the information processing system according to the second embodiment.
11 is a diagram showing an example of a display screen displayed on the terminal device in the linkage processing.
12 is a diagram showing an example of a display screen displayed on the communicator PC in the cooperative processing.

Hereinafter, embodiment of this invention is described using drawing.

[First Embodiment]

[Configuration]

[System configuration]

1 is a diagram showing a system configuration of the information processing system 1 according to the present invention.

The information processing system 1 constitutes a proxy call center system that integrates and processes call center tasks of multiple companies. In the information processing system 1, a virtual operator VO is used for voice inquiries from customers, and a virtual operator responds to a formal inquiry in which an answer is prepared in advance. In addition, in the information processing system 1, when a virtual operator executes a process (replenishment processing) to supplement information for generating a reply to a query in which a reply is not prepared in advance, a reply can be generated , Answer by the virtual operator. In addition, in the information processing system 1, inquiries that require correspondence by a human communicator are connected to the call centers of each company and processed.

In addition, in the information processing system 1, a standardizable inquiry and a response are added in sequence to a corresponding standard inquiry by the virtual operator, based on the history performed by the human communicator.

In addition, in the information processing system 1, a virtual operator raises a topic for a customer who has inquired, and conducts a conversation to present recommendation information to the customer based on the information identified through the conversation.

As shown in Fig. 1, the information processing system 1 includes a VO server 10, an IVR server 20, a plurality of computer telephony integration (CTI) servers 30, and an IP-PBX (Intenet Protocol). It comprises a Private Branch eXchange (40) server, a screen linkage server (50), a CRM (Customer Relationship Management) server (60), and a terminal device (70). In addition, the terminal device 70 and the IVR server 20 are configured to be able to communicate through the public telephone network 80A, and the terminal device 70, the VO server 10, and the CRM server 60 are such as the Internet. They are communicatively connected to each other via the public communication network 80B. In addition, the VO server 10 and the CTI server 30 are connected to each other so that they can communicate with each other through a VPN (Virtual Private Network) 80C. In addition, the CTI server 30 transmits and receives IP voice (voice communication using the Internet protocol) with the IVR server 20 via the IP-PBX server 40.

The VO server 10 responds to inquiries from the terminal device 70 (smartphone or PC, etc.) mounted with a browser through a public communication network and responds to the primary reception using a virtual operator. Specifically, the VO server 10 recognizes the voice transmitted from the terminal device 70 and responds by voice according to the content of the utterance. In addition, the VO server 10 records the conversation between the customer who made the inquiry and the virtual operator, and performs machine learning based on the recorded conversation. As a result of the machine learning, the VO server 10 changes the method of speech recognition or the method of interpreting the customer's intention. In addition, the VO server 10 normalizes the response to the same inquiry to the response by the virtual operator based on the history of the inquiry corresponding to the communicator.

The IVR server 20 responds to an inquiry from the terminal device 70 through the public telephone network by automatic voice response.

The CTI server 30 is installed in each of a plurality of companies (herein referred to as company A to company C), and inquiries beyond the corresponding range of primary acceptance by the VO server 10 are human in each company. Lead to acceptance by the communicator. Specifically, a communicator PC used by the communicator is connected to the CTI server 30, and a communicator telephone is connected through the PBX. Further, in the CTI server 30, documents provided to customers, URL and PDF files, text data of FAQs, and the like are stored as a database. In addition, when the CTI server 30 accepts the communication, the display of the screen is controlled through the screen linkage server 50 that links the screen of the terminal device 70 used by the customer with the screen of the communicator PC. . In addition, the communicator and the customer can make a call by transmitting/receiving IP voice or voice communication through the public telephone network 80A.

The IP-PBX server 40 converts IP voice transmitted from the IVR server 20 or the terminal device 70 to voice in a telephone line. The screen-associated server 50 controls the terminal device 70 used by the customer and the terminal device for a communicator used by the communicator, and the terminal device for a communicator used by the screen and the communicator of the terminal device 70 used by the customer. Link the screens.

The CRM server 60 stores information on customer attributes, such as purchase history, behavior history, or preference information. In addition, these information memorize|stored by the CRM server 60 is hereafter called enemy "personal information". The CRM server 60 refers to personal information according to a request from the VO server 10 or the CTI server 30, and provides information that enables a suitable response to each customer individually.

The terminal device 70 is a terminal device 70 used by customers who inquire to a call center system constituted by the information processing system 1, and is configured of, for example, a smartphone, a PC, or a mobile phone. Among the terminal devices 70, devices capable of displaying a guide on the screen by the VO server 10, such as a smartphone or a PC, in addition to the voice response by the virtual operator, displays the guide by the screen. In addition, a voice response by a virtual operator is performed on the terminal device 70 that cannot display a guide by a screen.

The terminal device 70 communicates page data such as text and images or IP voice through the public communication network 80B. In addition, it is also possible to use, as the terminal device 70, voice communication on the telephone line through the public telephone network 80A. The terminal device 70 communicates an inquiry to the VO server 10 or the IVR server 20, and simultaneously displays or outputs a screen of a response transmitted from the VO server 10 or the IVR server 20. .

[Hardware configuration]

Next, the hardware configuration of the server and the terminal device 70 in the information processing system 1 will be described.

[Server hardware configuration]

In the information processing system 1, the VO server 10, the IVR server 20, the CTI server 30, the IP-PBX server 40, the screen linkage server 50, and the CRM server 60 are servers. It is configured by installing software that realizes the functions of each server on a computer.

2 is a diagram showing a hardware configuration of the server computer 800 constituting each server. As shown in FIG. 2, the server computer 800 constituting each server includes a central processing unit (CPU) 811, a read only memory (ROM) 812, and a random access memory (RAM) 813. ), a bus 814, an input unit 815, an output unit 816, a storage unit 817, a communication unit 818, and a drive 819.

The CPU 811 executes various processes (processes for realizing the functions of each server) according to programs recorded in the ROM 812 or programs loaded into the RAM 813 from the storage unit 817. . In the RAM 813, data necessary for the CPU 811 to perform various processes are also stored.

The CPU 811, the ROM 812, and the RAM 813 are connected to each other via the bus 814. An input unit 815, an output unit 816, a storage unit 817, a communication unit 818, and a drive 819 are connected to the bus 814.

The input unit 815 is composed of various buttons and the like, and inputs various information according to an instruction operation. The output unit 816 is composed of a display or a speaker, and outputs an image or audio. The storage unit 817 is composed of a hard disk, dynamic random access memory (DRAM), or the like, and stores various data managed by each server. The communication unit 818 controls communication performed with other devices through a network.

A removable medium 821 made of a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory is mounted on the drive 819. The program read from the removable medium 821 by the drive 819 is installed in the storage unit 817 as necessary.

[Hardware configuration of the terminal device 70]

In the information processing system 1, the terminal device 70 is constituted by a terminal device 70 equipped with a browser such as a smartphone or a personal computer (PC), or a mobile phone or a fixed phone capable of voice communication.

3 is a diagram showing a hardware configuration of the terminal device 70. In addition, FIG. 3 shows a hardware configuration of the smartphone as an example of the terminal device 70.

As shown in FIG. 3, the terminal device 70 includes a CPU 911, a ROM 912, a RAM 913, a bus 914, a touch manipulation input unit 915, and an output unit ( 916, an audio input unit 917, an imaging unit 918, a storage unit 919, a communication unit 920, and a drive 921.

The CPU 911 performs various processing according to a program recorded in the ROM 912 or a program loaded from the storage unit 919 to the RAM 913 (process for realizing the function of the terminal device 70) Run it. In the RAM 913, data necessary for the CPU 911 to execute various processes are also stored.

The CPU 911, the ROM 912, and the RAM 913 are connected to each other via the bus 914. A touch operation input unit 915, an output unit 916, an audio input unit 917, an imaging unit 918, a storage unit 919, a communication unit 920, and a drive 921 are connected to the bus 914. have.

The touch manipulation input unit 915 is configured by, for example, a capacitive or resistive position input sensor stacked on a display area of the output unit 916 to detect coordinates of a position where the touch manipulation is performed. Here, the touch operation refers to an operation of contact or proximity of an object (such as a user's finger or a touch pen) to the touch operation input unit 915. The output unit 916 is composed of a display, a speaker, and the like, and outputs an image or audio. That is, in the present embodiment, the touch panel is configured by the display of the touch manipulation input unit 915 and the output unit 916.

The audio input unit 917 is configured by, for example, a microphone, and converts the input audio into an analog electric signal. In addition, the analog signal of audio is subjected to A/D (Analog/Digital) conversion processing, and is converted into a digital signal representing audio data. The subject of execution of this A/D conversion process is not particularly limited, and may be a CPU 911 or dedicated hardware (not shown), but in the present embodiment, for convenience of explanation, it is assumed that the audio input unit 917 is used. That is, in the present embodiment, the voice input unit 917 outputs voice data based on the input voice to the CPU 911 or the like.

Although not illustrated, the imaging unit 918 includes an optical lens unit and an image sensor. The optical lens unit is composed of a lens that condenses light, for example, a focus lens, etc. in order to photograph an object. Further, in addition to the focus lens, an optical lens unit may be provided with a zoom lens that freely changes the focal length within a certain range. The optical lens unit is also provided with a peripheral circuit that adjusts setting parameters such as focus, exposure, and white balance as necessary.

The image sensor is composed of a photoelectric conversion element or an analog front end (AFE). The photoelectric conversion element is composed of, for example, a complementary metal oxide semiconductor (CMOS) type photoelectric conversion element or the like. In this embodiment, the Bayer array color filter is attached to the imaging element. The imaging element performs photoelectric conversion (imaging) of the optical signal on the subject, which has been incident and accumulated therebetween, at regular intervals, and sequentially supplies the resulting analog electrical signal to the AFE. The AFE performs various signal processing such as A/D conversion processing on the analog electrical signal. A digital signal is generated by various signal processing, and is output as an output signal of the imaging unit 918. The data of the image displayed by the output signal of the imaging unit 918 is hereinafter referred to as the enemy "photographed image data". The captured image data is stored in the RAM 913 or the storage unit 919 according to the instruction of the CPU 911.

The storage unit 919 is composed of a storage device such as DRAM, and stores various data. The communication unit 920 controls communication performed with other devices through a network. A removable medium 931 is mounted on the drive 921 as necessary. The program read from the removable medium 931 by the drive 921 is installed in the storage unit 919 as necessary. In addition, the removable media 931 can also store various data stored in the storage unit 919 as in the storage unit 919.

[Functional configuration]

Next, the main functional configuration realized in the information processing system 1 will be described. 4 is a block diagram showing a main functional configuration realized in the information processing system 1.

As shown in Fig. 4, the information processing system 1 is a database, the VO server 10, a character database 151, a personal database 152, a scenario database 153, and a FAQ search database 154, a speech synthesis database 155, a marketing template database 156, a VOC database 157, a topic raising database 158, a communicator history database 159, and a conversation history database 160 It is equipped with. In addition, the CTI server 30 is provided with a document database 351 and a FAQ text database 352. In addition, the CRM server 60 is provided with a CRM personal database 651 and a meta database 652. In addition, in FIG. 4, the database is abbreviated as "DB" and is described.

The character database 151 stores data representing various expressions and behaviors of the virtual operator. The personal database 152 stores, for each user (customer) specified by the user information, user-specific information such as past inquiry content or a part of CRM personal data obtained from the CRM server 60.

The scenario database 153 stores scenarios (patterns of responses) in which responses or conversations are formally answered by a virtual operator. The FAQ search database 154 stores Frequently Asked Questions (FAQ) data. The speech synthesis database 155 stores speech data for synthesizing sentences as speech.

The marketing template database 156 stores data of a template that presumes various marketing or questioning questions. The VOC database 157 stores customer opinions or requests (VOC: Voice Of Customer) transmitted to the information processing system 1 as voice or text data.

The topic raising database 158 stores data of topics to be raised according to the contents of the inquiry to the customer who made the inquiry. The communicator history database 159 stores data of the history of the conversation between the customer and the communicator.

The conversation history database 160 stores data of a conversation history between the customer and the virtual operator. The document database 351 stores data such as documents, URLs or PDFs that the communicator presents to the customer. The FAQ text database 352 stores text data of the FAQ that the communicator presents to the customer.

The CRM personal database 651 stores user personal attribute information or related information such as purchase history, behavior history, or preference information. The meta database 652 stores and stores purchase information, behavior information, SNS information, and preference information of a large number of users (ie, as metadata).

4, the information processing system 1, as a function of the IVR server 20, includes an inquiry voice receiving unit 201, a response requesting unit 202, a reply receiving unit 203, It is equipped with a reply voice transmission unit 204. In addition, as a function of the VO server 10, the IP communication unit 101, the utterance start unit 102, the user information acquisition unit 103, the voice recognition unit 104, the emotion analysis unit 105 and , Language discrimination section 106, intention/status analysis section 107, conversation processing section 108, speech synthesis section 109, screen control section 110, supplementary processing section 111, communicator call A unit 112, a machine learning processing unit 113, a formalization processing unit 114, and a communicator support unit 115 are provided. In addition, as a function of the CTI server 30, a response management unit 301 and a screen control unit 302 are provided.

The IP communication unit 101 transmits the IP voice transmitted from the terminal device 70 to the IP-PBX server 40. The utterance start unit 102 detects the silent portion of the input voice to start utterance. The user information acquisition unit 103 is based on information (IP address, user ID and password, terminal number, etc.) of the accessed terminal, and user information (user-specific number, etc.) that uniquely identifies the user who is the user of the terminal. To acquire.

The speech recognition unit 104 speech-recognizes the speech input from the speech start unit 102 and converts the speech into a character string. The emotion analysis unit 105 analyzes the emotion of the speaker based on the utterance and silence portions input from the utterance starting unit 102 and outputs data quantifying the emotion. For example, the emotion analysis unit 105 calculates a numerical value for each parameter, such as joy, anger, sadness, joy, calmness, or impatience, with respect to the emotion of the speaker, based on the speed or the size of the accent spoken by the speaker do.

The language discrimination unit 106 determines the spoken language based on the character string input from the speech recognition unit 104. For example, the language discrimination unit 106 assumes a plurality of languages to which the information processing system 1 corresponds, and among the results recognized by the speech recognition unit 104, the accuracy of the analysis result is the highest. It is determined that it is a spoken language.

The intention/state analysis unit 107 determines the intention of the talker and the current conversational state based on the character strings in which the language has been determined by the language determination unit 106 and numerical data of emotions input from the emotion analysis unit 105. Interpret. Specifically, the intention/state analysis unit 107 determines the content of a character string (that is, a character string in which the speaker's voice is text) by which the language is determined by the language discrimination unit 106, and the intention of the talker's conversation Interpret (the meaning of the utterance). At this time, the intention/status analysis unit 107 is based on the user information acquired by the user information acquisition unit 103, the personal data and personal database 152 of the user stored in the CRM server 60 The intention of the talker's conversation is interpreted by referring to the contents of the user's past inquiry stored in the file. For example, when a speech-recognized talker contains a word having plural homonyms, the intention/status interpreter 107 refers to the content of the talker's past inquiry or purchase history. The meaning of the word is narrowed and interpreted as the most likely meaning. Thereby, it is possible to reliably determine the intention of the inquiry according to the speaker (customer) who made the inquiry, and to reply to the inquiry. In addition, the intention/state analysis unit 107 determines that the interpretation of the intention is wrong when the utterance repeats the same question with a strong feeling of "anger" about the answer of the virtual operator when interpreting the intention of the utterance And it is possible to change the interpretation of the intention.

In addition, the intention/state analysis unit 107 interprets the prerequisites of the presenter's current conversation as elements of the current conversational state from the flow of the conversation up to the present after the inquiry is made. For example, if the talker specifies a specific condition in one utterance (for example, Shinjuku is designated as the region), the specific condition in which the utterance spoken in the next utterance has already been specified. It is interpreted as a premise.

Further, the intention/state analysis unit 107 interprets the type of the terminal device 70 used by the talker as an element of the current conversation state. For example, the intention/state analysis unit 107 interprets whether the terminal device 70 used by the talker is a terminal device 70 capable of being guided by a screen display as an element of the current conversation state. . When the terminal device 70 used by the talker is the terminal device 70 that is not capable of being guided by the screen display, the intention/state analysis unit 107 selects a conversation that does not use an image.

The intention/state analysis unit 107 searches the FAQ search database 154 based on the speaker's intention and the current conversation state interpreted as described above to determine whether or not the conversation content can be determined (first judgment) ). As a result of the first determination, if the content of the conversation can be determined, the intention/state analysis unit 107 determines the content of the conversation about the utterance. For example, when the talker utters, "What is a delicious restaurant in Shinjuku?", the intention/status analysis unit 107 determines the content of the conversation to answer the popular restaurant in Shinjuku. On the other hand, when the conversation content cannot be determined, the intention/status analysis unit 107 requests the supplementary processing unit 111 to supplement the processing (to be described later) of the information for determining the conversation content. Then, the intention/state analysis unit 107 determines whether or not the content of the conversation can be determined based on the result of the supplementary processing (second determination). Specifically, is it possible to search the FAQ search database 154 using the results of the supplementary processing and determine the content of the conversation, or whether information to determine the content of the conversation is obtained from information sources other than the FAQ search database 154? Is judged. As a result of the second judgment, when it is possible to determine the content of the conversation, the intention/state analysis unit 107 determines the content of the conversation about the utterance. On the other hand, if the content of the conversation cannot be determined, the intention/state analysis unit 107 calls the communicator of the company in question through the communicator call unit 112. In addition, when the inquiry is made by chatting (conversation by text only) from the terminal device 70, the character string transmitted by the chat is directly input to the intention/status analysis unit 107, and thereafter, the inquiry by voice Similarly, the content of the conversation is decided. However, in this case, the response to the terminal device 70 is performed by chatting using a character string.

The dialogue processing unit 108 refers to the scenario database 153 based on the contents of the dialogue interpreted by the intention/state analysis unit 107 to determine the dialogue to respond to the speaker. For example, if the conversation processing unit 108 is determined by the intention/status analysis unit 107 to be a conversation content for answering a popular restaurant in Shinjuku, according to the scenario stored in the scenario database 153, "Shinjuku A popular restaurant located within ○○m from the station is… Select the dialogue called 」.

In addition, the conversation processing unit 108 stores the utterance of the utterance and the utterances responding to the utterance in the conversation history database 160 as a conversation history by correlating the user information of the utterance with the terminal device 70 of the utterance.

The speech synthesis unit 109 speech-synthesizes the dialogue determined by the conversation processing unit 108 and transmits the speech to the talker's terminal device 70. The screen control unit 110 refers to the character database 151 based on the conversation content analyzed by the intention/state analysis unit 107 and displays character data of the virtual operator displayed on the talker's terminal device 70. Decide. Specifically, the screen control unit 110 selects character data of facial expressions and actions corresponding to the content of the conversation interpreted by the intention/state analysis unit 107, and the virtual operator communicates with the facial expressions and actions. Generate data. Then, the screen control unit 110 transmits the generated screen data to the talker's terminal device 70.

The supplementary processing unit 111 performs supplementary processing of information for determining the content of a conversation when a request for supplementary processing is requested from the intention/state analysis unit 107. Specifically, when there are a plurality of candidates for answering an inquiry, the supplementary processing unit 111 transmits a question to supplement the contents of the inquiry to the terminal device 70, or the answer to the inquiry is the FAQ search database 154 If not included in the search, information related to the inquiry content is searched from an information source (communication history database 159 or the Internet) other than the FAQ search database 154. Then, the supplement processing unit 111 outputs an answer to the question or a search result to the intention/status analysis unit 107 as a result of the supplement processing.

The communicator call unit 112 calls the communicator of the inquiring company when a request is made to call the communicator from the intention/state analysis unit 107. In addition, when an operation for invoking the communicator in the terminal device 70 (for example, pressing down the "call" button in FIG. 6A, described later), etc., the communicator invoking unit 112 communicates with the communicator. Execute processing to call. At this time, the communicator call unit 112 may notify the communicator of the talk rate or emotional numerical data of the utterance acquired in a conversation with the virtual operator in addition to the contents of the customer's inquiry as the utterance.

The machine learning processing unit 113 refers to the conversation history stored in the conversation history database 160, and determines whether an inappropriate response is being made in the conversation history. For example, the machine learning processing unit 113 determines whether or not there is a situation in which the talker has denied the response of the virtual operator or instructed to connect to a communicator other than the virtual operator in the conversation history. Then, the machine learning processing unit 113 notifies the system manager PC used by the system administrator that the conversation history determined that an inappropriate response has been performed.

Contents of speech recognition (conversion from speech to text) in the conversation history when the machine history processing unit 113 determines that the conversation history determined by the system administrator PC is inappropriate as the inappropriate response, or As the accuracy of the speaker's interpretation of the intention (the result of discrimination of the contents of the string) is low, it is fed back to the processing by the speech recognition unit 104 and the intention/state analysis unit 107.

The shaping processing unit 114 refers to the communicator history database 159 by executing a teasing management process, and if a plurality of responses to the same inquiry are being made by the communicator, the inquiry and the response to the same are normalized, It is added to the FAQ search database 154. In addition, at this time, the standardized inquiry and answer may be notified to the system administrator PC and, if approved, addition to the FAQ search database 154 may be performed.

Here, in the present embodiment, it is assumed that the inquiries and answers to the communicators in the call centers of multiple companies are referred to as one unit, and that the standardization is performed according to whether or not multiple answers to the same inquiry are being made. This makes it possible to standardize the same inquiries and replies in multiple companies, making it possible to increase the frequency of formalizing inquiries and replies compared to the case where a single company operates a call center. However, even if it refers to the inquiry and the reply to the communicator in the call center for each company, and the company answers the inquiry to the same inquiry according to whether or not there are multiple answers to the same inquiry, even if the company answers the inquiry do.

The communicator support unit 115, when the communicator responds to an inquiry by the customer, voice-recognizes the conversation by the voice recognition unit 104, and based on text data as a result of voice recognition, information supporting the communicator Provides For example, the communicator support unit 115 displays information on the "NISA" on the communicator PC when the content of the inquiry by the customer is "NISA (small tax exemption system)". In addition, the communicator support unit 115 may analyze the conversation between the customer and the communicator by the emotion analysis unit 105 and display the emotion change of the customer on the communicator PC based on the data quantifying the emotion. .

The inquiry voice reception unit 201 of the IVR server 20 receives the voice transmitted from the terminal device 70 and outputs it to the response request unit 202. The response request unit 202 transmits the voice data input from the inquiry voice receiving unit 201 to the VO server 10 to request a response voice. The reply receiving unit 203 receives the voice data of the reply transmitted from the VO server 10. The response voice transmission unit 204 transmits the voice data of the response received by the response reception unit 203 as voice to the terminal device 70.

The response management unit 301 of the CTI server 30 manages the inquiry connected to the communicator by correlating the customer with the correspondent communicator. The screen control unit 302 controls the screen of the terminal device 70 through the screen association server 50 according to the operation of the communicator.

[action]

Next, the operation of the information processing system 1 will be described. 5 is a flowchart showing the flow of the reception processing executed by the information processing system 1.

The reception processing starts with the activation of the information processing system 1. 6 is a diagram showing an example of a display screen displayed on the terminal device 70 in the reception process, FIG. 6(A) is an example of a reception initial screen, and FIG. 6(B) is a display screen in a formal response Yes, Fig. 6(C) is an example of a communicator screen, and Fig. 6(D) is a diagram showing an example of a display screen in VOC collection. Hereinafter, the flow of the acceptance process will be described with reference to FIG.

In step S1, the voice recognition unit 104 of the VO server 10 determines whether or not there is an inquiry from the terminal device 70. If there is no inquiry from the terminal device 70, it is determined as NO in step S1, and the processing in step S1 is repeated. On the other hand, when there is an inquiry from the terminal device 70, it is determined as YES in step S1, and the process proceeds to step S2.

In step S2, the screen control unit 110 causes the terminal device 70 to display the initial acceptance screen (see FIG. 6(A)). In step S3, the utterance start unit 102 starts the utterance, and the voice recognition unit 104 recognizes the voice input from the utterance start unit 102, and the emotion analysis unit 105 starts the utterance. Emotion analysis is performed on the voice input from (102). In step S4, the language discrimination unit 106 determines the spoken language.

In step S5, the intention/state analysis unit 107 analyzes the speaker's intention and the current conversation state. In step S6, the intention/state analysis unit 107 searches the FAQ search database 154 based on the intention of the speaker analyzed in step S5 and the content of the utterance specified from the current conversation state. In step S7, the intention/state analysis unit 107 determines whether or not the content of the conversation can be determined (that is, whether or not there is an answer in the FAQ search database 154).

If the conversation content can be determined, it is determined as YES in step S7, and the process proceeds to step S8. On the other hand, if the conversation content cannot be determined, it is determined as NO in step S7, and the process proceeds to step S10. In step S8, the intention/status analysis unit 107 determines the content of the conversation by the answer retrieved from the FAQ search database 154, and performs a response (standard response) by the conversation processing unit 108 (Fig. 6). (B)).

In step S9, the intention/state analysis unit 107 determines whether or not the response to the inquiry has been completed. If the response to the inquiry has not been completed, the process proceeds to step S3. On the other hand, when the response to the inquiry is completed, the process proceeds to step S18.

In step S10, the intention/state analysis unit 107 requests the supplementary processing unit 111 for the supplementary processing. Specifically, when there are a plurality of candidates for answering an inquiry, the supplementary processing unit 111 sends a question to supplement the inquiry content to the terminal device 70, or an answer to the inquiry is a FAQ search database 154 If not included in the search, information related to the inquiry content is searched from an information source (communication history database 159 or the Internet) other than the FAQ search database 154.

In step S11, the intention/state analysis unit 107 determines whether or not the content of the conversation can be determined as a result of the supplementary processing. As a result of the supplementary processing, if the conversation content can be determined, it is determined as YES in step S11, and the processing proceeds to step S12. On the other hand, if the conversation content cannot be determined as a result of the supplementary processing, it is determined as NO in step S11, and the process proceeds to step S13.

In step S12, the intention/state analysis unit 107 determines the content of the conversation using the result of the supplementary processing, and responds by the conversation processing unit 108. After step S12, the process proceeds to step S9.

In step S13, the intention/state analysis unit 107 calls the communicator of the company as the inquiry destination through the communicator call unit 112. At this time, the screen control unit 110 may respond to the virtual operator with the effect of invoking the communicator on the screen of the terminal device 70. In addition, when an operation for invoking the communicator is performed in the terminal device 70, the interrupt processing jumps to the processing in step S13, whereby the communicator is called.

In step S14, the screen control unit 110 of the CTI server 30 sets a picture of the communicator on the screen of the terminal device 70 as a background, and a communicator screen in which conversation contents are textualized by speech recognition (FIG. 6 ( C) reference). By displaying the communicator screen, the customer can feel the person (communicator) who is responding closer, making it possible to perform inquiries and responses more smoothly. In step S15, the voice recognition unit 104 voice-recognizes the conversation between the customer and the communicator.

In step S16, the intention/state analysis unit 107 determines whether or not the conversation between the customer and the communicator has ended. If the conversation between the customer and the communicator has not ended, it is determined as NO in step S16, and the process proceeds to step S14. On the other hand, when the conversation between the customer and the communicator is finished, it is determined as YES in step S16, and the process proceeds to step S17.

In step S17, the intention/state analysis unit 107 executes VOC collection processing (see Fig. 6(D)). In step S18, the conversation processing unit 108 stores a series of conversation history for the inquiry in the conversation history database 160. After step S18, the acceptance processing is repeated until an end instruction is issued.

By such processing, in the case where an inquiry is made from the terminal device 70 in the information processing system 1, when an answer is prepared in the FAQ search database 154 with reference to the FAQ search database 154, a virtual operator The formal response by is done. On the other hand, if an answer is not prepared in the FAQ search database 154, the VO server 10 performs supplemental processing of the information for determining the content of the conversation, and uses the result of the supplemental processing to find the FAQ search database ( 154) is searched for an answer, or processing is performed to obtain information from an information source other than the FAQ search database 154, and a virtual operator responds to the inquiry. In addition, even if the result of the supplementary processing is used, if the response to the inquiry cannot be made by the virtual operator, the communicator is called and the communicator responds. Therefore, in the information processing system 1, the response by the virtual operator is attempted as much as possible, and the frequency of calling the communicator is reduced. Therefore, it becomes possible to make the task of accepting inquiries from customers more efficient.

Next, VOC collection processing executed as a sub-flow in step 17 of the reception processing will be described. 7 is a flowchart showing the flow of VOC collection processing. When the VOC collection process is started, in step S21, the intention/status analysis unit 107 refers to the marketing template database 156 to obtain a questionnaire template regarding the communicator's response, and is displayed by the screen control unit 110. The questionnaire is displayed on the screen of the terminal device 70.

In step S22, the intention/state analysis unit 107 receives a response to the questionnaire. In addition, the answer to the questionnaire can be input by the terminal device 70 in addition to voice input.

In step S23, the intention/state analysis unit 107 refers to the topic raising database 158, and raises the topic according to the inquiry content to the terminal device 70. In step S24, the voice recognition unit 104 recognizes a voice in response to a topic transmitted from the terminal device 70.

In step S25, the intention/state analysis unit 107 searches for a conversation for presenting cell information from the topic raising database 158 based on keywords included in the response obtained by speech recognition. In addition, the cell information is advertisement information intended for an up cell or a cross cell. In the case of presenting the cell information, it is possible to present the cell information limited to the company of the customer's inquiry or to present the cell information crossing a plurality of companies using the information processing system 1.

In step S26, the intention/state analysis unit 107 determines whether or not the conversation for presenting the cell information can be searched. When the conversation for presenting the cell information can be searched, it is determined as YES in step S26, and the process proceeds to step S27. On the other hand, if the conversation for presenting the cell information cannot be searched, it is determined as NO in step S26, and the process proceeds to step S28.

In step S27, the intention/state analysis unit 107 performs a conversation for presenting the cell information, and presents the cell information to the customer. After step S27, the process proceeds to step S29. In step S28, the intention/state analysis unit 107 searches the terminal device 70 for a topic to be searched and raised from the topic raising database 158.

In step S29, the intention/state analysis unit 107 determines whether or not it matches the end conditions of the VOC collection processing. VOC collection processing termination conditions may be defined when a customer instructs the end of VOC collection processing or when a customer requests product guidance according to cell information. If it does not match the end conditions of the VOC collection process, it is determined as NO in step S29, and the process proceeds to step S24. On the other hand, if it is consistent with the end condition of the VOC collection process, it is determined as YES in step S29, and the VOC collection process ends.

By such processing, the virtual operator and the customer perform related conversation (comparative conversation) other than the inquiry itself, and the cell information according to the content of the conversation is presented to the customer. Therefore, it is possible to present the cell information matched with the customer's state to the customer who made the inquiry due to the occurrence of an event or the like. Therefore, more effective marketing can be performed using the call center service.

Next, the teasing management process will be described. 8 is a flowchart showing the flow of the animation management process executed by the information processing system 1. The play management process is executed as a batch process or the like at a preset time.

In step S31, the shaping processing unit 114 refers to the communicator history database 159. In step S32, the shaping processing unit 114 analyzes the history of the communicator history database 159.

In step S33, the formalization processing unit 114 determines whether or not there is a conversation that can be formalized with the history of the communicator history database 159. Whether or not there is a conversation that can be formalized can be determined by, for example, whether or not a response to the same inquiry is performed a predetermined number of times or more. If there is no conversation that can be formalized with the history of the communicator history database 159, it is determined as NO in step S33, and the teasing management process ends. On the other hand, if there is a conversation that can be formalized with the history of the communicator history database 159, it is determined as YES in step S33, and the process proceeds to step S34.

In step S34, the formalization processing unit 114 adds the formalized inquiry and answer to the FAQ search database 154. After step S34, the teasing management process ends.

As a result of such processing, when the same inquiry is made multiple times with respect to an unstructured inquiry for which the communicator has responded, the inquiry and response are added to the FAQ search database 154 as a standardizable inquiry. Then, when the same inquiry is made later, a response by a virtual operator, not a communicator, is performed. Therefore, as the information processing system 1 is operated, the range of inquiries that formally responds is expanded, so that the efficiency of the call center work using the information processing system 1 can be improved.

[Second Embodiment]

Next, a second embodiment of the present invention will be described.

[Configuration]

9 is a block diagram showing a main functional configuration realized in the information processing system 1 according to the second embodiment. The information processing system 1 according to the second embodiment, in the information processing system 1 according to the first embodiment, the linkage server 90 for realizing linkage between the VO server 10 and the IVR server 20 ). In Fig. 9, the function of each server shown in the first embodiment is omitted.

The linkage server 90 includes an linkage processing unit 901 that executes processing for linkage between the VO server 10 and the IVR server 20. The cooperative processing unit 901 responds to the IVR server 20 in response to a customer using the terminal device 70 performing an operation to call the communicator when the primary reception counter is performed by the VO server 10. ) To make a call. Then, the linkage processing unit 901 automatically inquires by voice according to the contents of the conversation with the customer so far, and responds to the automatic voice response provided by the IVR server 20, until the call of the communicator. Conversation proceeds automatically. By such processing, even when the VO server 10 and the CTI server 30 cannot directly link, the communicator calls from the VO server 10 by using the automatic voice response function by the IVR server 20. It becomes possible to do it. In addition, it is possible that the VO server 10 is provided with the function of the linked server 90.

[action]

Next, the operation will be described. The information processing system 1 which concerns on 2nd Embodiment performs the reception processing shown in FIG. 6 similarly to 1st Embodiment. Then, in the case where an operation for invoking the communicator is performed in the terminal device 70, the cooperative process is executed as a sub-flow of step S13. 10 is a flowchart showing the flow of the cooperative processing executed by the information processing system 1 according to the second embodiment. 11 is a diagram showing an example of a display screen displayed on the terminal device 70 in the cooperative process, FIG. 11(A) is an example of an input screen of a telephone number, and FIG. 11(B) is a list of contacts of the communicator 11C is a diagram showing an example of a screen displayed during the call of the communicator. 12 is a diagram showing an example of a display screen displayed on the communicator PC in the cooperative process, FIG. 12(A) is an example of a conversation history search screen, and FIG. 12(B) is an example of a conversation history confirmation screen, FIG. 12 (C) is an example of a screen when receiving a call, and FIG. 12(D) is a diagram showing an example of a screen after answering is completed. Hereinafter, the flow of the linkage processing will be described with reference to Figs.

In step S41, the intention/status analysis unit 107 of the VO server 10 responds by prompting the customer to input his/her telephone number by the conversation processing unit 108 (see Fig. 11(A)). At this time, the telephone number input by the customer may be either input by voice or input by a character string. Thereby, the customer's own telephone number is obtained from the VO server 10. In step S42, the communicator call unit 112 obtains a list of the contact information of the communicator, which is supposed to be a customer's inquiry, from the association processing unit 901 of the association server 90, and transmits it to the terminal device 70 (FIG. 11). (B)). In FIG. 11(B), three contact information likely to be a customer's contact point and buttons for displaying all contact information are displayed. The contact information of the communicator can be provided from the CTI server 30 to the linkage processing unit 901, or the linkage processing unit 901 can collect and acquire information from a general web server. In addition, as the contact information of the communicator, it is possible to indicate the department name and telephone number of the inquiries or only the department names of the inquiries. In addition, in step S42, the data of the conversation history between the customer and the virtual operator so far is transmitted to the linkage server 90.

In step S43, the communicator call unit 112 acquires the contact information selected by the terminal device 70. In step S44, the communicator call unit 112 requests the connection processing unit 901 to send a call from the terminal device 70 to the selected contact. In addition, in step S44, the data of the conversation history between the customer and the virtual operator so far is transmitted to the linkage server 90. In step S45, the linkage processing unit 901 makes an outgoing call to the IVR server 20, and automatically inquires by voice in accordance with the content of the previous conversation with the customer. At this time, the cooperative processing unit 901 automatically proceeds to the conversation until the communicator is called while responding to the automatic voice response provided by the IVR server 20. Further, a message to the effect that the communicator is being called is notified to the terminal device 70 of the customer by the virtual operator (see Fig. 11(C)).

In step S46, the IVR server 20 calls the communicator as a result of the response to the automatic conversation by the linkage processing unit 901. In step S47, the response management unit 301 of the CTI server 30 executes a call answering process executed when there is a call to the communicator. In the call answering process, on the communicator PC, a conversation history search screen for retrieving the contents of the inquiry is displayed (see Fig. 12(A)), and the communicator inputs the ID notified from the linkage server 90. Then, in the conversation history confirmation screen, the conversation history (data of the conversation history managed by the connection server 90) with the customer who made the call is displayed as a string, and the button for the communicator to send a call to the customer (Answer call button) is displayed (see Fig. 12(B)). In step S48, the coordination processing unit 901 shifts to the "waiting to receive a call" state in which the communicator is ready to make a call in response to the operator operating the call answering button. At this time, in the communicator PC, a screen is displayed when receiving a call to input a result after the call answer button operation (see FIG. 12(C)).

In step S49, the linkage processing unit 901 sends a call from the communicator telephone to the telephone number input by the customer through the IP-PBX server 40. In step S50, when the customer answers the phone, the communication between the communicator and the customer is started. After step S50, the processing returns to the reception processing. In addition, after the communication between the communicator and the customer is ended, the communicator, on the screen when answering the call, displays “Available for calls”, “Don't answer the phone”, “Outside the phone”, “Missed answering machine”, and “Call out” Enter the result of contacting the customer who said "Middle" or "Incorrect phone number". When input by the communicator is performed, a screen is displayed after completion of the response displayed after completion of the response by the communicator to the customer (see FIG. 12(D)).

By such processing, when the customer requests the call of the communicator, the conversation processing unit 901 of the association server 90 proceeds to input a conversation for automatic voice response in the IVR server 20, The communicator via the IVR server 20 can be called. Therefore, even when the VO server 10 and the CTI server 30 cannot directly connect, it is possible to make a call from the VO server 10 to the communicator using the automatic voice response function by the IVR server 20. Becomes

[Application Example 1]

In the above-described embodiment, the character or costume of the virtual operator can be different depending on the company in question. For example, in the character database, you can remember multiple types of characters with different personalities, and if the inquiries are technical, make the virtual operator a character that is familiar with technology, and if it is about banking, use the virtual operator as a character in hard clothes. Can. This makes it possible to feel the atmosphere of the virtual operator suitable for the contents of the inquiry to the customer who made the inquiry.

[Application Example 2]

In the above-described embodiment, the character or costume of the virtual operator can be set to be different depending on the customer who made the inquiry. For example, if a plurality of types of characters having different personalities are stored in the character database, and the customer who inquires is interested in animation, the virtual operator can be used as the animation character. As a result, a response from the call center can be performed by a virtual operator who gives a favorable impression to the customer, thereby improving customer satisfaction.

[Application Example 3]

In the above-described embodiment, when the customer who inquired performs the same procedure for multiple companies, such as address change, it may be possible to perform procedures for multiple companies at the same time by the reception of the virtual operator. Thereby, the customer can reduce the trouble of performing the same procedure multiple times.

[Application Example 4]

In the above-described embodiment, procedures other than the contents inquired by the customer, and procedures related to the contents inquired may be proposed by the virtual operator, and these procedures may be performed at once. Thereby, the customer can reduce the possibility of omission of necessary procedures.

As described above, the information processing system 1 according to the present embodiment includes a terminal device 70 and a VO server 10 as a reception server. The VO server 10 includes a speech recognition unit 104, an intention/status analysis unit 107, and a communicator call unit 112. The voice recognition unit 104 accepts an inquiry from the terminal device 70. The intention/state analysis unit 107 determines whether an answer to an inquiry can be searched from the FAQ search database 154 in which a plurality of questions and answers prepared in advance are stored. In addition, when it is determined that the answer to the inquiry cannot be retrieved, the intention/status analysis unit 107 executes a supplementary process to supplement the information for answering the inquiry, and based on the result of the supplementary process , Create a reply to the inquiry. In addition, the intention/state analysis unit 107 determines whether or not an answer to the inquiry can be generated. Further, the intention/state analysis unit 107 transmits the searched response or the generated response when it is determined that the answer to the inquiry can be searched and when it is determined that the answer to the inquiry can be generated. The communicator call unit 112 executes the call processing of the communicator corresponding to the inquiry when it is determined by the intention/state analysis unit 107 that an answer to the inquiry cannot be generated. In this way, when an inquiry is made from the terminal device 70, when the answer is prepared in the FAQ search database 154 with reference to the FAQ search database 154, a formal response is performed. On the other hand, when an answer is not prepared in the FAQ search database 154, a supplementary process of information for answering an inquiry is executed, and an answer is generated based on the result of the supplementary process. Further, even if supplemental processing is performed, if a reply to the inquiry cannot be generated, the communicator is called, and a response by the communicator is performed. Therefore, in the information processing system 1, an automatic response by the system is attempted as much as possible, and the frequency of calling the communicator decreases. Therefore, it becomes possible to make the task of accepting inquiries from customers more efficient.

In addition, the intention/state analysis unit 107 transmits a query to the terminal device 70 to supplement the contents of the inquiry as a supplementary process. Thereby, when the answer to the inquiry cannot be narrowed, a question to narrow the answer can be asked.

In addition, the intention/state analysis unit 107, as a supplementary process, retrieves information regarding the contents of the inquiry from information sources other than the FAQ search database 154. In this way, even if an answer to the inquiry is not prepared, information can be obtained from the searchable information source and answered.

In addition, the voice recognition unit 104 accepts an inquiry from the terminal device 70 by automatic response using voice recognition. In addition, the intention/status analysis unit 107 transmits the searched response or the generated response as a response by voice data. Thereby, it is possible to automatically respond to an inquiry from the terminal device by voice.

In addition, the VO server 10 includes a shaping processing unit 114. The shaping processing unit 114 generates a combination of shaping questions and replies based on the history of the communicator's responses. Accordingly, when the same inquiry is made multiple times with respect to an unstructured inquiry for which the communicator has responded, it is considered that the inquiry and response are questions that can be formalized. Therefore, the more the information processing system 1 is operated, the more the range of inquiries that formally responds is expanded, so that the efficiency of work using the information processing system 1 can be improved.

Further, the communicator call unit 112 executes a call process to call the communicator of the organization corresponding to the inquiry among the communicators installed corresponding to each of the plurality of organizations. The voice recognition unit 104 representatively accepts inquiries from a plurality of organizations. The shaping processing unit 114 generates a combination of an inquiry and a reply that can be standardized for a plurality of organizations in common, based on the history of communication by the communicator with respect to each organization. In this way, in the case where the correspondence service to the inquiry in multiple organizations is integrated, the same inquiry and response in multiple organizations can be formalized, so that the response service to the inquiry in a single organization is operated. Compared to the case, it is possible to increase the frequency with which the questions and answers can be formalized.

In addition, the VO server 10 includes a personal database 152. The personal database 152 manages personal information relating to individual customers making inquiries. The intention/state analysis unit 107 refers to the personal information and determines whether or not an answer to the inquiry can be searched from the FAQ search database 154. Thereby, it is possible to reliably determine the intention of the inquiry according to the customer who made the inquiry, and to reply to the inquiry.

In addition, the intention/state analysis unit 107 provides a topic related to the content of the inquiry as a voice. In addition, the intention/status analysis unit 107 voice-recognizes the response from the user of the terminal device 70 to the topic, and provides recommendation information (cell information, etc.) related to the response content as voice. In this way, the system and the customer perform related conversations (mischievous conversations) other than the inquiry itself, and recommendation information according to the content of the conversation is presented to the customer. For this reason, it is possible to present recommendation information consistent with the customer's status to the customer who made an inquiry due to an event or the like. Therefore, more effective marketing can be performed by using the correspondence task to the inquiry.

Further, the information processing system 1 includes an IVR server 20 as an automatic voice response server, and an associated server 90. The IVR server 20 responds to the preset inquiry from the terminal device 70 by an automatic voice response, and executes the call processing of the communicator other than the preset inquiry. When the communicator server 90 makes a call to the communicator from the VO server 10, based on the inquiry and response made between the VO server 10 and the terminal device 70, the associated server 90 is connected to the IVR server 20. An inquiry is made, conversation is automatically performed until the call processing of the communicator is performed, and the call of the communicator is executed. Thereby, when the customer requests the call of the communicator, the communication server 90 proceeds to input the conversation for the automatic voice response in the IVR server 20, and passes through the IVR server 20. The communicator can be called. Therefore, it becomes possible to make the call of the communicator from the VO server 10 using the automatic voice response function by the IVR server 20.

In addition, this invention is not limited to the above-mentioned embodiment, The deformation|transformation, improvement, etc. in the range which can achieve the objective of this invention are contained in this invention. For example, in the above embodiment, the organization represented by the information processing system 1 is described as a company, but is not limited thereto. That is, the present invention can be applied to a school, an administrative institution, a medical institution, etc., in addition to a company, as long as it is an organization that responds to inquiries.

In addition, in the above-described embodiment, the configuration of each server is shown as an example, and if the functions of the respective servers are provided in the information processing system 1 as a whole, the functions of a plurality of servers may be integrated or The function can be divided.

The above-described series of processes can be executed by hardware or can be executed by software. In other words, the functional configuration of FIG. 4 is merely an example and is not particularly limited. That is, it is sufficient if the information processing system 1 is provided with a function capable of executing the above-described series of processes as a whole, and it is not particularly limited to the example of FIG. 4 which function block is used to realize this function. Further, one functional block may be composed of a single hardware component, a single software component, or a combination thereof.

When a series of processing is executed by software, a program constituting the software is installed from a computer or the like from a network or recording medium. The computer may be a computer mounted on dedicated hardware. In addition, the computer may be a computer capable of executing various functions by installing various programs, for example, a general-purpose personal computer.

The recording medium including such a program is not only constituted by the removable media 821 of FIG. 2 distributed separately from the apparatus body to provide a program to the user, but also provided to the user in a state pre-mounted in the apparatus body It consists of a recording medium or the like. The removable media 821 is made of, for example, a magnetic disk (including a floppy disk), an optical disk, or a magneto-optical disk. The optical disk is composed of, for example, a CD-ROM (Compact Disk-Read Only Memory), a DVD (Digital Versatile Disk), or the like. The magneto-optical disk is composed of MD (Mini-Disk) or the like. In addition, the recording medium provided to the user in a state pre-mounted in the apparatus main body is constituted of, for example, a DRAM 812 included in the ROM 812 of FIG. 2 or a storage unit 817 of FIG. 2 in which a program is recorded. do.

In addition, in the present specification, steps for describing a program recorded on a recording medium include not only processes performed in time series according to the order, but also processes executed in parallel or individually even if not necessarily processed in time series. Is to do. In addition, in this specification, the term system is meant to mean an overall device composed of a plurality of devices or a plurality of means.

1: Information processing system
10: VO server
20: IVR server
30: CTI Server
40: IP-PBX server
50: screen link server
60: CRM server
70: terminal device
80A: public telephone network
80B: public telecommunication network
80C: VPN
90: linked server
800: server computer
811, 911: CPU
812, 912: ROM
813, 913: RAM
814, 914: Bus
815: input unit
816, 916: output
817, 919: memory
818, 920: Ministry of Communications
819, 921: drive
821, 931: removable media
915: touch operation input
917: voice input unit
918: imaging unit
151: character database
152 personal database
153: scenario database
154: FAQ search database
155: speech synthesis database
156: Marketing template database
157: VOC database
158: Topic database
159: Communicator history database
160: conversation history database
351: Document database
352: FAQ text database
651: CRM personal database
652: meta database
201: Inquiry Voice Receiver
202: response request unit
203: reply receiving unit
204: reply voice transmitter
101: IP communication unit
102: beginning of ignition
103: user information acquisition unit
104: speech recognition unit
105: emotional analysis unit
106: language discrimination unit
107: intention/state analysis unit
108: conversation processing unit
109: speech synthesis unit
110, 302: screen control
111: supplement processing unit
112: communicator calling unit
113: machine learning processing unit
114: shaping processing unit
115: Communicator support department
301: Response management unit
901: linked processing unit

Claims (12)

An information processing system comprising a terminal device for sending an inquiry and a reception server for accepting the inquiry sent from the terminal device,
The reception server,
Reception means for accepting the inquiry from the terminal device;
First determination means for determining whether an answer to the inquiry can be retrieved from a response database in which a plurality of questions and answers prepared in advance are stored;
When it is determined by the first determination means that the answer to the inquiry cannot be retrieved, a supplementary process is performed to supplement the information for performing the answer to the inquiry, and based on the result of the supplementary process, Answer generating means for generating a reply to the inquiry;
Second determination means for determining whether or not an answer to the inquiry could be generated by the answer generation means;
When it is determined by the first determination means that an answer to the inquiry can be retrieved, and when it is determined by the second determination means that an answer to the inquiry can be generated, by the first determination means Reply means for transmitting the searched reply or the reply generated by the second determination means;
When it is determined by the second determination means that an answer to the inquiry cannot be generated, a calling means for executing a call processing of a communicator corresponding to the inquiry
Including,
The reply generating means is, as the supplementary processing, information about the contents of the inquiry from an information source other than the response database, a data of the history of the conversation between the customer who made the inquiry in the past and the communicator, and the Internet. Search,
The answer generation means supplements the contents of the inquiry to select, for the terminal device, a response more consistent with the inquirer's intention, as a supplementary process when a plurality of responses to the inquiry are selectable. Send a question,
The information processing system,
An automatic voice response server that responds to a preset inquiry from the terminal device by an automatic voice response and executes a call processing by a communicator other than the preset inquiry;
When making a call to the communicator from the reception server, when making an inquiry to the automatic voice response server based on the inquiry and response made between the reception server and the terminal device, when the communicator's call processing is performed Up to the automatic voice answering server, a linkage server that automatically conducts a conversation and executes a call to the communicator
Information processing system characterized in that it further comprises. delete According to claim 1,
The reception means accepts the inquiry from the terminal device by automatic response using voice recognition,
And the answering means transmits the reply searched by the first judgment means or the reply generated by the second judgment means as a response by voice data. According to claim 1,
The reception server,
And standardization processing means for generating a combination of standardized inquiries and responses based on the history of the communicator's responses. According to claim 4,
The calling means executes the call processing to call the communicator of the organization corresponding to the inquiry among the communicators installed corresponding to each of the plurality of organizations,
The receiving means is representatively accepting inquiries to the plurality of organizations,
The standardization processing means corresponds to a corresponding inquiry in which multiple answers to the same inquiry are performed by the communicator installed corresponding to each of a plurality of organizations based on the history of the communication by the communicator with respect to the inquiry to each organization. An information processing system characterized by generating an answer to an inquiry in a combination of an inquiry and an answer that can be standardized in common to the plurality of organizations. According to claim 1,
The reception server includes personal information management means for managing personal information regarding the purchase history, behavior history, and preference information of the individual customer making the inquiry,
The first judging means determines whether or not an answer to the inquiry can be retrieved from the response database with reference to the personal information regarding the purchase history, behavior history, and preference information of the individual customer making the inquiry. Information processing system, characterized in that. According to claim 1,
The reception server,
A topic providing means for providing a topic related to the content of the inquiry as a voice;
And information recommendation means for speech recognition of a response from the user of the terminal device to the topic and providing recommendation information related to the response content as voice. An information processing system comprising a terminal device for sending an inquiry and a reception server for accepting the inquiry sent from the terminal device,
The reception server,
Reception means for accepting the inquiry from the terminal device;
First determination means for determining whether an answer to the inquiry can be retrieved from a response database in which a plurality of questions and answers prepared in advance are stored;
When it is determined by the first determination means that the answer to the inquiry cannot be retrieved, a supplementary process is performed to supplement the information for performing the answer to the inquiry, and based on the result of the supplementary process, Answer generating means for generating a reply to the inquiry;
Second determination means for determining whether or not an answer to the inquiry could be generated by the answer generation means;
When it is determined by the first determination means that an answer to the inquiry can be retrieved, and when it is determined by the second determination means that an answer to the inquiry can be generated, by the first determination means Reply means for transmitting the searched reply or the reply generated by the second determination means;
When it is determined by the second determination means that an answer to the inquiry cannot be generated, a calling means for executing a call processing of a communicator corresponding to the inquiry
Equipped with,
The information processing system,
An automatic voice response server that responds to a preset inquiry from the terminal device by an automatic voice response and executes a call processing by a communicator other than the preset inquiry;
When making a call to the communicator from the reception server, when making an inquiry to the automatic voice response server based on the inquiry and response made between the reception server and the terminal device, when the communicator's call processing is performed The information processing system further comprises a link server that automatically proceeds to a conversation between the automatic voice response server and executes a call to the communicator. According to claim 1,
The reception server,
Selecting character data of a virtual operator that accepts and responds by the accepting means by the accepting means, generates data of a screen spoken by the selected virtual operator, and transmits the generated screen data to the terminal device An information processing system, comprising screen control means. delete An information processing method executed in an information processing system comprising a terminal device for sending an inquiry and a reception server for accepting the inquiry sent from the terminal device,
A reception step of accepting the inquiry from the terminal device;
A first judging step of judging whether an answer to the inquiry can be retrieved from a response database in which a plurality of questions and answers prepared in advance are stored;
In the first judging step, when it is determined that the answer to the inquiry cannot be retrieved, a supplementary process is performed to supplement information for performing the answer to the inquiry, and based on the result of the supplementary process, A reply generation step of generating a reply to the inquiry;
In the answer generation step, a second determination step of determining whether a response to the inquiry could be generated,
In the first determination step, when it is determined in the first determination step that an answer to the inquiry can be retrieved, and in the second determination step it is determined that an answer to the inquiry can be generated. A response step of transmitting the searched response or the response generated in the second determination step;
And in the second determination step, if it is determined that an answer to the inquiry cannot be generated, a calling step of executing a call processing of a communicator corresponding to the inquiry,
The reply generation step includes, as the supplementary processing, information on the contents of the inquiry from an information source other than the response database, a data of the history of the conversation between the customer who made the inquiry in the past and the communicator, and the Internet. Search,
The reply generation step is to supplement the contents of the inquiry in order to select an answer more consistent with the inquirer's intention with respect to the terminal device, as a supplementary process when a plurality of answers to the inquiry are selectable. Send a question,
The information processing system,
An automatic voice response server that responds to a preset inquiry from the terminal device by an automatic voice response and also performs a call processing by a communicator for other than the preset inquiry;
When a communicator is called from the reception server, an inquiry to the automatic voice response server is made based on the inquiry and response made between the reception server and the terminal device, and communication processing of the communicator is performed. Up to the automatic voice response server, the conversation server automatically proceeds with the conversation and executes the call of the communicator.
Information processing method characterized in that it further comprises. delete

Images