JP2009182433A - Information provision system, information providing device, information providing method, and information provision program of call center - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information provision system, an information providing device, an information providing method and an information provision program of a call center, which can extract the need of a client and provide information suitable for the need. <P>SOLUTION: A call center system 1 for providing a client with guide information comprises: a category information storage section storing a keyword associated with the guide information; a provision information storage section for determining whether a keyword appears in speech communication between the operator of a call center and the client or not and storing provision data indicating a call center for providing guide information; and an information provision server 12 for selecting a call center associated with the keyword based on the provision information when a keyword has appeared. Furthermore, an automatic connector 7 for connecting a client telephone 5 to the selected call center is provided. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a call center information providing system, information providing apparatus, information providing method, and information providing program.

The call center is a department that specializes in telephone answering operations such as inquiries about products from customers and receiving orders for products. In general, systems built in call centers include private branch exchanges (PBX Private Branch eXchange), automatic distribution of incoming calls (A
A CD Automatic Call Distribution) device, a terminal used by an operator, and various databases (see, for example, Patent Document 1). The database stores customer information that stores customer attributes, purchase history, etc., history information that stores past response history, etc., and these information are used to support telephone response, marketing, It is used for adjusting the number of operators.
JP 2007-228273 A

  However, even though the call with the customer is an opportunity to extract the potential needs of the customer, it is difficult to extract the needs of the customer during the call because the operator concentrates on the service. Even if the operator can extract the customer's needs, the extracted needs cannot be utilized.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a call center information providing system, information providing apparatus, and information capable of extracting customer needs and providing information that meets the needs. It is to provide a providing method and an information providing program.

  The present invention provides a call center information providing system for providing guidance information to a customer, keyword storage means for storing a keyword associated with the guidance information, and whether the keyword appears in a call between a call center operator and the customer. Determination means for determining whether or not, call center information storage means storing call center information indicating a call center providing the guidance information, and a call center associated with the call center information and the keyword when the keyword appears. Selecting means for selecting; and connecting means for connecting the customer's telephone to the selected call center.

  According to the present invention, it is possible to determine whether or not a preset keyword has been uttered during a call, and to grasp the potential needs of the customer from the appearing keyword. Further, since the customer's telephone is connected to a call center that provides guidance information associated with the keyword, the needs extracted from the call with the customer can be actively utilized.

  The call center information providing system further includes priority determining means for determining the priority of the guidance information based on the appearance status of the keyword in a call, and the selecting means has a higher priority than the priority determining means. The call center that provides the guidance information determined to be selected is selected.

According to the present invention, since the customer's telephone is connected to the call center that provides the high-priority guide information, the guide information that meets the customer's needs can be provided.
In the call center information providing system, the call center information providing system further includes emotion detection means for generating an emotion detection value indicating the customer's emotion based on a voice feature amount of the customer's call voice, and the priority determination means is configured such that the keyword is spoken. The priority of the guidance information corresponding to the keyword is determined using the emotion detection value of the customer at the time.

  According to this invention, the customer's emotion when the keyword is spoken is detected, and the priority of the guidance information is determined according to the detection result. For this reason, since presence / absence of potential needs can be determined based on the type of detected customer emotion, potential needs can be more reliably extracted.

  In the call center information providing system, the keyword belongs to a category, the guidance information is associated with the category, and the priority determination unit uses the emotion detection value when the keyword is spoken as the category. The priority is determined according to the total sum of the detected emotion values for each category.

  According to this invention, for each category to which the keyword belongs, the emotion detection values are summed to determine the priority of the guidance information. Therefore, information that is highly necessary for the customer can be determined by objectively determining the strength of the customer's interest.

The call center information providing system further includes introduction information providing means for transmitting introduction information about the customer to the selected call center.
According to this, when the customer's telephone is connected to the selected call center from the call center that has been in communication with the customer, introduction information about the customer is transmitted. Therefore, the call center operator newly connected to the customer's telephone can smoothly provide the guidance information based on the introduction information.

  According to the present invention, in a call center information providing apparatus for providing guidance information to a customer, keyword storage means for storing a keyword associated with the guidance information, and whether the keyword has appeared in a call between a call center operator and the customer Determination means for determining whether or not, call center information storage means storing call center information indicating a call center providing the guidance information, and a call center associated with the call center information and the keyword when the keyword appears. Selecting means for selecting.

  According to the present invention, it is possible to determine whether or not a preset keyword has been uttered during a call, and to grasp the potential needs of the customer from the appearing keyword. Further, since the customer's telephone is connected to a call center that provides guidance information associated with the keyword, the needs extracted from the call with the customer can be actively utilized.

  The present invention provides a call center information providing method using a control means for providing guidance information to a customer, wherein the control means is stored in a keyword storage means, and a keyword associated with the guidance information is defined between the call center operator and the A step of determining whether or not the call has appeared in a call with a customer; and when the keyword appears, the call center stored in the call center information storage means and indicating the call center providing the guidance information is associated with the keyword Selecting a call center.

  According to this method, it is possible to determine whether or not a preset keyword is spoken during a call, and to grasp the potential needs of the customer based on the keyword that has appeared. Further, since the customer's telephone is connected to a call center that provides guidance information associated with the keyword, the needs extracted from the call with the customer can be actively utilized.

  The present invention provides a call center information providing program using a control means for providing guidance information to a customer, wherein the control means is stored in a keyword storage means, and a keyword associated with the guidance information is defined between the call center operator and the A determination means for determining whether or not an appearance has occurred in a call with a customer, and when the keyword appears, the call center information stored in the call center information storage means and indicating the call center providing the guidance information is associated with the keyword. Function as a selection means for selecting a given call center.

  According to the present invention, it is possible to determine whether a preset keyword is spoken during a call according to the information providing program, and to grasp the potential needs of the customer based on the appearing keyword. Further, since the customer's telephone is connected to a call center that provides guidance information associated with the keyword, the needs extracted from the call with the customer can be actively utilized.

  Hereinafter, an embodiment embodying the present invention will be described with reference to FIGS. FIG. 1 is a network diagram in which call center systems 1A, 1B,... As a plurality of information providing systems are connected. In addition, when not distinguishing each call center system 1A, 1B ..., it demonstrates as the call center system 1 only.

  Each call center system 1 according to the present embodiment supports a response operation such as inquiry and ordering of products from a customer, and also provides an information providing service. In this information providing service, customer needs are extracted based on calls with customers. When the needs are extracted, a call center that provides guidance information that meets the customer's needs is selected from each of the affiliated call centers, and the customer's telephone is connected to the selected call center system 1.

  As shown in FIG. 1, each call center system 1A, 1B,... Includes PBXs 2A, 2B,..., Management servers 3A, 3B, ..., operator terminals 4A, 4B, and database (D / B) servers 8A, 8B・ Each has .., Management servers 3A, 3B,..., Operator terminals 4A, 4B... And database (D / B) servers 8A, 8B. The operator terminal 4 and the D / B server 8 will be described.

  The PBX 2, the management server 3, and the operator terminal 4 are connected via a LAN (Local Area Network) 9 so that various data can be transmitted and received. The PBX 2 is connected to a customer's telephone (hereinafter referred to as customer telephone 5) via the public telephone line network N1. The PBX 2 is connected to each telephone 6 used by the operator. The operation unit of each telephone 6 is provided with a login button, a hold button, and a preparing button (not shown). The login button is a button for receiving an incoming call on the telephone 6. The hold button is a button for holding the connected telephone, and the ready button is turned on when the telephone is being prepared.

  Moreover, PBX2A, 2B of each call center is connected by the automatic connection apparatus 7 as a connection means. The automatic connection device 7 is a PBX or the like, and connects the customer telephone 5 connected to the telephone 6A of the call center system 1A to the PBX 2B of another call center system 1B based on the operation of the operator. The PBX 2 connects the transfer telephone to the telephone 6B of the call center system 1B.

The operator terminal 4 includes a CPU, RAM, ROM, communication interface (I
/ F) and the like, and a plurality of units are connected to the LAN 9. The operator terminal 4 is connected to an input device I such as a mouse and a keyboard and a display D. The operator operates the input device I while talking to the customer using a headset (not shown) that receives the voice from the telephone 6 and inputs the call voice of the operator.

  The management server 3 has a CTI (Computer Telephony Integration) function for connecting the telephone 6 to the call center system 1 and controls incoming calls to the call center and outgoing calls from the call center. Specifically, the management server 3 has an ACD function, an IVR (Interactive Voice Response) function, and the like, and controls the PBX 2. The ACD function is a function for automatically allocating a call to a call center to a waiting operator. The IVR function is a voice automatic response function.

  The D / B server 8 includes a customer information storage unit, a response history information storage unit, and a guidance voice storage unit (not shown). The customer information stored in the customer information storage unit is information related to a customer who has called the call center in the past or information related to a customer registered in advance. The management server 3 reads the corresponding customer information from the D / B server 8 based on the customer identification data such as the name inputted by the operator's operation or the customer identification data inputted via the customer telephone 5, and the operator terminal 4 is displayed on the display D.

  The response history information storage unit stores response history data for each incoming call. The response history data is the response contents input by the operator using the operator terminal 4 and includes an incoming call number, an incoming call date, a call start time and an end time, an operator ID, a product model number, a customer ID, a holding history, and the like. is doing.

  Further, each of the call center systems 1A, 1B,... Includes speech recognition devices 10A, 10B, .. emotion detection devices 11A, 11B as emotion detection means, information provision servers 12A, 12B,. Servers 13A, 13B,. If the voice recognition devices 10A, 10B,..., The emotion detection devices 11A, 11B,..., The information providing servers 12A, 12B,. The detection device 11, the information providing server 12, and the web server 13 will be described.

  In the present embodiment, the voice recognition device 10, the emotion detection device 11, the information providing server 12, and the web server 13 are different devices. However, a plurality of these devices may be used as one device, or all devices. May be one server.

As shown in FIG. 2, the speech recognition apparatus 10 includes a speech recognition processing unit 15, a speech conversion unit 16, a communication I / F 17, a recognition information storage unit 18, and a text data storage unit 21.
The voice recognition processing unit 15 is stored from a CPU, RAM, ROM, etc., or an LSI for emotion recognition, and performs voice recognition processing during a call between a customer and an operator according to a voice recognition program stored in a storage unit (not shown). . The recognition information storage unit 18 stores recognition information 19 used for voice recognition processing. As the recognition information 19, for example, an acoustic model in which a feature amount of a speech and a phoneme are associated, a recognition dictionary in which tens of thousands to hundreds of thousands of words associated with a phoneme string are stored, a probability of being located at the beginning / end of a sentence In addition, there are various data such as a connection probability between consecutive words and a language model modeling a dependency relationship.

The voice recognition processing unit 15 inputs a voice signal from the telephone 6 or the like according to a voice recognition program, and A / D converts the voice signal by the voice conversion unit 16 to make a data format capable of voice recognition. Note that the voice conversion unit 16 can be omitted when voice data in a format that can be voice-recognized from the outside, such as the telephone 6, can be acquired. Further, the voice recognition processing unit 15 converts the converted voice data into character data using the recognition information 19. A known method can be used for this treatment. For example, the speech recognition processing unit 15 calculates speech waveform features using speech data, and selects phonemes by comparing the feature amounts with an acoustic model. Also, each phoneme string and the recognition dictionary are collated to select a recognition candidate word. Furthermore, the speech recognition processing unit 15 calculates the probability of connection relation using the language model, and determines consistency. When the recognition result is confirmed, the text is stored in association with the time when the text is spoken.

  Further, the voice recognition processing unit 15 assigns identification data to the voice recognition text to generate text data 22, and temporarily stores the text data 22 in the text data storage unit 21.

  As shown in FIG. 3, the emotion detection device 11 includes an emotion detection unit 25, a voice conversion unit 26, a communication I / F 27, a detection information storage unit 28, and an emotion detection information storage unit 29. The emotion detection unit 25 has a CPU, RAM, ROM, etc., or an LSI for emotion recognition, etc., and calculates an emotion detection value obtained from the feature amount of the call voice according to an emotion detection program stored in a storage unit (not shown). To do.

  The emotion detection device 11 receives a voice signal from the telephone 6 or the like, and A / D converts the voice signal by the voice conversion unit 26 to make a data format capable of detecting emotion. Note that the voice conversion unit 26 can be omitted when voice data in a data format capable of emotion detection can be acquired from an external device such as the telephone 6. The emotion detection unit 25 performs emotion detection during a call between a customer and an operator. In the present embodiment, for each type of emotion such as “joy” and “excitement”, a value indicating the strength of emotion is specified, but in order to extract customer needs other than “joy” and “excitement” The strength of the emotion may be detected.

  A known method can be used as the emotion detection method. For example, the emotion detection unit 25 calculates the number of phonemes in the detection target section based on the generated voice data, and acquires the number of phonemes as the tempo. The detection target section may be a section divided every predetermined time, or may be a word, a phrase, or a sentence separation recognized by the speech recognition apparatus 10 or the like. Alternatively, based on the waveform of audio data, an area including a predetermined pattern may be set as the detection target section.

  Moreover, the emotion detection part 25 acquires the audio | voice intensity | strength for every detection object area. Also, the intensity of the voice is detected, and the voice inflection is calculated based on the intensity change pattern. Furthermore, the emotion detection unit 25 may detect physical feature quantities of speech such as a scale, a pitch, a melody, and a frequency.

  The detection information storage unit 28 stores detection information 30 used for emotion detection processing. As the detection information 30, data such as a voice characteristic pattern in a joy state and a voice characteristic pattern in a state of excitement are stored. The emotion detection unit 25 compares the fluctuation pattern of the voice feature amount based on each voice signal with a previously stored pattern, and identifies the current emotion strength. In the present embodiment, values such as “1L”, “2L”,... Are set as values representing the strength of each emotion such as “joy” and “excitement”. One of the values is specified as emotional strength.

  The emotion detection unit 25 specifies the strength of the emotion detected for each detection target section as an emotion detection value. That is, for one detection target section, a detection value of “joy” and a detection value of “excitement” are specified, and the detection value is associated with the detection target section. Then, identification data is assigned to each emotion detection value to generate emotion detection data 31, which is temporarily stored in the emotion detection information storage unit 29.

As shown in FIG. 4, the information providing server 12 includes a CPU 35, a RAM 36, a ROM 37, a communication I / F 38, a category information storage unit 39 as a keyword storage unit, a provision information storage unit 41 as a call center information storage unit, and a record information storage. Part 50 is provided. CPU3
5 performs various controls using the RAM 36 as a work area based on an information providing program stored in the ROM 37 or a storage unit (not shown). The CPU 35 corresponds to a determination unit, a selection unit, a priority determination unit, an introduction information providing unit, and a control unit.

  The information providing servers 12A and 12B of the respective call center systems 1A and 1B are connected to routers R and firewalls (not shown) installed in the respective call center systems 1A and 1B via a network N2 such as a WAN. Necessary data can be transmitted and received in both directions.

  Further, the CPU 35 is connected to the LAN 9 via the communication I / F 38, acquires the text data 22 from the voice recognition device 10, and acquires the emotion detection data 31 from the emotion detection device 11.

  The category information storage unit 39 stores category data 40 in which predetermined language units such as words, phrases, or sentences are categorized. As shown in FIG. 6, the category data 40 has a category 40B and a keyword 40A belonging to the category 40B. The CPU 35 determines whether or not the keyword 40A is included in the text data 22 acquired from the speech recognition apparatus 10, and when the keyword 40A is included in the text data 22, as shown in FIG. The keyword 40A is extracted. Then, the category 40B to which the keyword 40A belongs is determined based on the category data 40.

  Further, the CPU 35 determines the priority of the guidance information provided to the customer based on the appearance status of the keyword 40A in the call. Specifically, the CPU 35 collates the acquired text data 22 and the emotion detection data 31 to acquire the emotion detection value 45 for the extracted keyword 40A. For example, when the detection target section 46 is divided every predetermined time, the time when the keyword 40A is uttered and the detection target section 46 of the emotion detection value 45 based on the time data associated with the text data 22. And the emotion detection value 45 corresponding to the keyword 40A is obtained. When the detection target section 46 associated with the emotion detection value 45 is divided for each word, the position of the keyword 40A in the text data 22 is compared with the order of the detection target section 46 from the start of the call. Then, the emotion detection value 45 corresponding to the keyword 40A is obtained.

  For example, as shown in FIG. 7, when “detection target section (1)” corresponds to the keyword 40A such as “mouse”, the emotion detection value 45 corresponding to “detection target section (1)” is “2L”. "Is identified as the emotion detection value 45 corresponding to" mouse ".

  When the emotion detection value 45 associated with the keyword 40A is acquired, the CPU 35 calculates the emotion detection value 47 for each category. That is, the CPU 35 adds the emotion detection value 47 of the keyword 40A belonging to each category 40B. For example, when keywords 40A such as “photograph” and “poster” belonging to the category “printer” 40B are extracted, the emotion detection value 45 associated with “photograph” is “6L” and “poster”. The emotion detection value 47 such as “13L” is calculated by adding the associated emotion detection value 45 “7L”.

  As a result, an emotion detection value 47 for each category is calculated. The emotion detection value 47 of the category 40B indicates the strength of emotion of “joy” and “excitement” when the customer speaks the keyword 40A belonging to the category 40B. That is, the higher the emotion detection value 47, the higher the priority for providing the information of the category 40B.

In this priority determination process, when emotion detection is not performed as in the present embodiment, for example, when the priority of the category 40B in which the number of appearances of the keyword 40A is high is increased, the following problems are predicted to occur. Is done. That is, when the customer is delighted or excited, the number of appearances of the keyword 40A belonging to a certain category 40B increases, and the priority of the category 40B to which the keyword 40A belongs can be increased, but also when the customer is angry. It is predicted that the keyword 40A belonging to a certain category 40B is uttered repeatedly. As a result, even when the customer is angry, the priority of the category 40B to which the keyword 40A belongs is set high.

  On the other hand, when statistics of the emotion detection value 45 for each category are not performed as in the present embodiment, it is predicted that the following problems will occur. For example, when one keyword 40A having the highest emotion detection value 45 of “joy” and “excitement” is selected from the emotion detection values 45 and the priority of the category 40B corresponding to the keyword 40A is increased, the customer Even when the voice inflection is changed accidentally or the strength is increased, a high priority is assigned to the category 40B belonging to the keyword 40A. Therefore, by categorizing the keyword 40A and detecting the emotion, the category 40B in which the customer is interested can be more reliably estimated.

  When the emotion detection value 47 for each category is calculated, the CPU 35 determines a category having a high priority based on the emotion detection value 47. Specifically, the category 40B having the highest emotion detection value 47 is selected from each category 40B.

  Next, the CPU 35 acquires a URL (Uniform Resource Locator) corresponding to the category 40 </ b> B having a high priority from the provision data 42 as call center information stored in the provision information storage unit 41. As schematically shown in FIG. 7, in the provided data 42, an operator 48 and a URL 49 are associated with each category. The business operator 48 is a partner company of the information providing service, and each business operator presets a keyword 40A for providing guidance information. The URL 49 indicates a place where web data for guiding the product or service of the business operator is stored. The business operator 48 may be omitted. In addition, the category 40B may be associated with the name of a file having guidance information instead of the URL 49.

  URL 49 indicates the hierarchy of the web server 13. The CPU 35 extracts the URL 49 of the category 40B having the highest emotion detection value 47. On the other hand, the operator terminal 4 transmits a transmission request for the URL 49 to the information providing server 12 every predetermined time. The information providing server 12 receives the transmission request and extracts the URL 49, and then transmits the URL 49 to the operator terminal 4. To do. Based on the received URL 49, the operator terminal 4 reads web data W (see FIG. 1) corresponding to the URL 49 from the web server 13.

  The web data W stored in the web server 13 is data prepared by the information provider service partner as described above, and the web data W associated with the “PC” category 40B is, for example, a personal computer. It has guidance information provided by a business that manufactures or sells. The web data associated with the category “children” 40B has guidance information provided by businesses that sell children's products and services. The category “printer” 40B includes guidance information provided by a business operator that manufactures or sells printers.

  When the operator terminal 4 acquires the web data W, the operator terminal 4 develops the web data W by a web browser and displays a guidance screen on the display D. This guidance screen displays guidance information related to products or services that are presumed to be of potential customer interest. For example, in the case where the priority of the category “printer” 40B is high, for example, texts, photos, etc. explaining a new model printer are displayed on the guidance screen.

  For example, when an operator of one call center looks at the guidance screen displayed on the display D and determines that there is a potential need for “printer”, for example, Alternatively, it is asked whether or not the customer telephone 5 is connected to a call center that handles services. If the customer approves, the operator presses the hold button, presses a predetermined button or number, etc., and transfers it to the PBX 2A. At this time, the operator may press the number for designating the call center to which the operator connects, or the information providing server 12A may store the data designating the call center of the business operator 48 associated with the guidance information having the highest priority. You may transmit to PBX2A via N2.

  At this time, the information providing server 12A sends the identification data of the call center and necessary data such as the customer name out of the customer information as introduction information to the new information providing server 12 of the connection destination via the network N2. Send. The data included in the introduction information may be only call center identification data or data other than the data described above.

  The PBX 2A determines a call center connected to the customer telephone 5 and transmits a connection request to the automatic connection device 7. The automatic connection device 7 connects the customer telephone 5 to the designated call center PBX 2B.

  The PBX 2B newly connected to the customer telephone 5 connects to the telephone 6B designated by the management server 3B. In addition, when the information providing server 12 newly connected to the customer telephone 5 receives the identification data of the call center of the transfer source and the customer information, another information is displayed on the display D of the operator terminal 4 corresponding to the transferred telephone 6B. A pop-up function displays a screen indicating that the call is being transferred from the call center.

The operator confirms the transfer notification screen, responds to the incoming call, and guides the product or service based on the fact that the transfer call is from another call center.
In addition, the record information storage unit 50 of the information providing server 12 stores record data 51 that indicates the number of times a call with a customer is transferred to another call center for each business operator. As shown in FIG. 5, the business record 48 and the number of connections 52 are stored in association with each other in the result data 51. When the information providing server 12 transmits the URL 49, the information providing server 12 adds “1” to the connection count 52 of the business operator 48 corresponding to the URL 49. In accordance with the record data 51, the call center may charge the above-described business operator 48 in cooperation. Note that the connection count 52 may be the number of times the operator actually provided guidance to the customer.

  Next, the processing procedure of the information providing server 12 will be described with reference to FIG. First, the CPU 35 of the information providing server 12 determines whether or not a call with a customer has been started (step S1). The method for determining the start of the call is not particularly limited. For example, when the login button of the telephone 6 is pressed, a start command is transmitted from the PBX 2 or the like to the information providing server 12, and the call starts when the CPU 35 receives the start command. You may make it judge that it is.

  If it is determined that the call has started (YES in step S1), the CPU 35 acquires the text data 22 from the speech recognition apparatus 10 at a predetermined timing (step S2). At this time, the text data 22 may be acquired every predetermined time. Moreover, CPU35 acquires the emotion detection data 31 from the emotion detection apparatus 11 (step S3).

Further, the information providing server 12 selects the category 40B having a high priority (Step S4). Specifically, the keyword 40A is extracted from the text data 22 using the category data 40 as described above. Further, the emotion detection value 45 corresponding to the keyword 40A is determined. Further, an emotion detection value 47 is calculated by summing the emotion detection value 45 for each category to which the keyword 40A belongs. Then, the category 40B having the highest emotion detection value 47 is selected.

  Further, the information providing server 12 transmits the URL 49 of the category 40B having a high priority to the operator terminal 4 using the provision data 42 (Step S5). That is, the information providing server 12 selects the URL 49 associated with the category 40B having the highest priority among the URLs 49 stored in the provision data 42. When the URL 49 transmission request is received from the operator terminal 4, the URL 49 is transmitted to the operator terminal 4 that has transmitted the transmission request.

  As a result, the operator terminal 4 reads the web data W from the web server 13 based on the received URL 49 and displays a screen including the guidance information selected based on the web data W on the display D. As described above, when a series of telephone reception is completed, the operator provides information displayed on the guidance screen to the customer and asks whether or not the call center that guides the product or service displayed on the screen can be connected. If the customer approves, the user presses the hold button on the telephone 6 and presses the number corresponding to the call center having the higher priority. When the PBX 2 receives a command based on the operation, the PBX 2 transmits a connection request for connecting the customer telephone 5 to the designated call center to the automatic connection device 7.

  When receiving the connection request, the automatic connection device 7 connects the customer telephone 5 and the designated call center PBX 2. The PBX 2 newly connected to the customer telephone 5 connects the customer telephone 5 to the operator telephone 6.

  Further, the information providing server 12 transmits necessary data among the identification data of the call center and the customer information corresponding to the customer in the call as introduction information to the information providing server 12B of the selected call center system 1B (step) S6).

  When the URL 49 is transmitted, the information providing server 12 updates the result data 51 (step S7). That is, the number of connections 52 of the business operator 48 corresponding to the transmitted URL 49 is extracted from the record data 51, and the number of connections 52 is incremented. When the record data 51 is updated, the process ends.

According to the above embodiment, the following effects can be obtained.
(1) In the above embodiment, the call center system 1 includes the information providing server 12 storing the keyword 40A preset by the provider providing the guidance information regarding the product or service, and the guidance information associated with the keyword 40A. And a web server 13 storing web data W. In addition, the voice recognition device 10 performs voice recognition on the call voice when the operator and the customer make a call, and the information providing server 12 determines whether or not the keyword 40A appears during the call. Further, when the keyword 40A appears during a call, a priority is given to the category 40B to which the keyword 40A belongs, and the category 40B having a higher priority is selected. For this reason, it is possible to grasp the potential needs of the customer by determining whether or not the keyword 40A preset by the business operator is uttered during a call. Further, the web data W associated with the category 40B is extracted, and the extracted web data W is displayed on the display D of the operator terminal 4. Further, the automatic connection device 7 connects the customer telephone 5 to the call center system 1 associated with the category 40B, and provides the customer with guidance information related to products and services. For this reason, the needs extracted from the call with the customer can be linked to the next sales activity.

(2) In the above embodiment, the information providing server 12 gives priority to the guidance information corresponding to the keyword 40A when the plurality of keywords 40A are extracted from the voice data during the call. Also, a call center that provides high priority guide information is selected from each guide information. For this reason, information which is highly necessary for the customer can be provided.

  (3) In the above-described embodiment, the call center system 1 further includes the emotion detection device 11 that detects a voice feature amount based on a customer's call voice and estimates a customer's emotion such as “joy” and “excitement”. In addition, the information providing server 12 determines the priority of the guidance information corresponding to the keyword 40A using the customer emotion detection value 45 when the keyword 40A is spoken. For this reason, it is possible to determine guidance information that is highly necessary for the customer by objectively determining the strength of the customer's emotion.

  (4) In the above embodiment, the information providing server 12 stores the keyword 40A for each category, and the web data including the guidance information is associated with the category. Further, the information providing server 12 adds the emotion detection value 45 of the keyword 40A for each category, and determines the priority of the guidance information corresponding to the category. For this reason, since the priority can be determined objectively in consideration of the appearance frequency of the keyword 40A for each category and the customer's emotion, it is possible to more reliably determine the guidance information that is highly necessary for the customer. Can do.

  (5) In the above embodiment, the information providing server 12 transmits the introduction information about the customer to the selected call center system 1B, so that the operator of the connected call center provides the guidance information smoothly based on the introduction information. can do.

In addition, you may change the said embodiment as follows.
In the above embodiment, the emotion detection value 45 of the keyword 40A is added for each category 40B, and the priority is determined according to the magnitude of the added emotion detection value 45, but the number of appearances of the keyword 40A for each category 40B And the priority may be determined based only on the number of appearances. That is, when the number of appearances of the keyword 40A belonging to the category 40B of “printer” is larger than the number of appearances of the keyword of the other category 40B, the guide information corresponding to “printer” is provided.

  In the above embodiment, voice recognition is performed on voice data generated by a customer's utterance. However, voice recognition may be performed on voice data in which an operator's voice and a customer's voice are mixed. Good. Or you may make it perform voice recognition with respect to the audio | voice data by an operator's utterance.

  In the above embodiment, the information providing server 12 determines the presence or absence of the keyword 40A with respect to the text data 22 that has been subjected to speech recognition. However, the information providing server 12 determines whether or not the keyword 40A is present during speech recognition. Also good. For example, a dictionary storing each keyword 40A may be created in the recognition information 19 stored in the recognition information storage unit 18 of the speech recognition apparatus 10, and the keyword 40A may be detected by referring to this dictionary. .

  In the above embodiment, the management server 3 plays the automatic guidance voice to the customer telephone 5 via the public telephone line network N1, but the product or service is displayed according to the screen displayed on the display D by the operator. You may make it perform guidance of.

  In the above embodiment, the web data as guide information is stored in the web server 13, but the URL 49 stored in the provided data 42 may be the URL of the homepage of the business operator.

Schematic of the call center system. The block diagram of a speech recognition apparatus. The block diagram of an emotion detection apparatus. The block diagram of an information provision server. Schematic diagram of actual data Schematic diagram of category data. Explanatory drawing of the process which selects guidance information. The flowchart which shows the process sequence of information provision service.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1,1A, 1B ... Call center system as information providing system, 6, 6A, 6B ... Telephone, 7 ... Automatic connection device as connection means, 11, 11A, 11B ... Emotion detection device as emotion detection means, 12 ... Information Information providing server as providing device, 35... Judging means, selecting means, priority determining means, connection request means, CPU as introduction information providing means and control means, 39... Category information storage section as keyword storing means, 40A. Keywords: 40B ... Category, 41 ... Provided information storage unit as call center information storage means, 42 ... Provided data as call center information, 45, 47 ... Emotion detection value.

Claims (8)

In a call center information provision system that provides guidance information to customers,
Keyword storage means for storing keywords associated with the guidance information;
A determination means for determining whether or not the keyword appears in a call between a call center operator and the customer;
Call center information storage means storing call center information indicating a call center providing the guidance information;
Selection means for selecting a call center associated with the call center information and the keyword when the keyword appears;
A call center information providing system comprising: connection means for connecting the customer's telephone to the selected call center. In the call center information provision system according to claim 1,
Priority determination means for determining the priority of the guidance information based on the appearance status of the keyword in a call,
The selection means includes
A call center information providing system that selects a call center that provides the guidance information determined to have a high priority by the priority determining means. In the call center information providing system according to claim 2,
Emotion detection means for generating an emotion detection value indicating the customer's emotion based on the voice feature amount of the customer's call voice,
The priority determination means includes
The call center information providing system, wherein the priority of the guidance information corresponding to the keyword is determined using the emotion detection value of the customer when the keyword is spoken. In the call center information providing system according to claim 3,
The keyword belongs to a category, the guidance information is associated with the category,
The priority determination means includes
The call center information providing system, wherein the emotion detection value when the keyword is uttered is added for each category, and the priority is determined according to the sum of the emotion detection values for each category. . In the call center information provision system according to any one of claims 1 to 4,
An information providing system further comprising introduction information providing means for transmitting introduction information about a customer to the selected call center. In a call center information providing device that provides guidance information to customers,
Keyword storage means for storing keywords associated with the guidance information;
A determination means for determining whether or not the keyword appears in a call between a call center operator and the customer;
Call center information storage means storing call center information indicating a call center providing the guidance information;
An information providing apparatus comprising: selection means for selecting a call center associated with the call center information and the keyword when the keyword appears. In a call center information providing method using a control means for providing guidance information to a customer,
The control means is
Determining whether a keyword stored in a keyword storage means and associated with the guidance information has appeared in a call between a call center operator and the customer;
Call center information indicating a call center that is stored in a call center information storage unit and provides the guidance information when the keyword appears, and selecting a call center associated with the keyword. Information provision method. In a call center information provision program using a control means for providing guidance information to customers,
The control means;
Determining means for determining whether or not a keyword stored in a keyword storage means and associated with the guidance information has appeared in a call between a call center operator and the customer;
When the keyword appears, the call center information is stored in a call center information storage unit and functions as a selection unit that selects call center information indicating a call center that provides the guidance information and a call center associated with the keyword. Information provision program.

Images