KR20110048675A - Call center counsel method and counsel system using voice recognition and tagging - Google Patents

Abstract

PURPOSE: A call center counseling method and counseling system which uses voice recognition and tagging are provided to search for knowledge related to a non-professional term by using voice recognition and tagging methods. CONSTITUTION: A recording database(10) records and stores conversation content between a client and a consultant. A voice recognizer(20) inputs a keyword which is extracted from the conversation content saved in the recording database. The voice recognizer recognizes the keyword about conversation content between the client and the consultant. A tagging unit(30) tags a related document which is searched by the consultant and generates tagging information. A display unit(70) displays the related document based on tagging information of the recognized keyword.

Description

Call center counseling method and counseling system using voice recognition and tagging

The present invention relates to a call center counseling method and counseling system using voice recognition and tagging, and more specifically, to a professional term used by a counselor using voice recognition and tagging technology, as well as an unprofessional and daily use that a customer understands and uses. Since the related terms are also keywords and related knowledge is searched, the present invention relates to a call center counseling method and counseling system using voice recognition and tagging that can provide information through a friendly search for everyday words used by customers.

In the general call center customer counseling system, when a customer requests information from a counselor by wire or wireless communication, the counselor retrieves the information from an information providing server including various DBs, and then retrieves the corresponding information retrieved from the customer. It will be sent as data.

At this time, each counselor keeps the main counseling knowledge in printed form or memorizes the main counseling contents to respond to customer inquiries.

In other words, for each call center, older counselors prepare counseling knowledge documents based on their know-how, key business knowledge and procedures, and updated information. Counselors are available for counseling.

In this type of call center information providing system, it is practically impossible to respond to the needs of customers through the printed form or memorization of the knowledge that is continuously accumulating in these days.

In addition, a voice recognition system is often introduced into a call center information providing system due to a burden of labor costs caused by a counselor directly handling information provision of all customer requests.

The information providing model through the voice recognition system has been mainly made using the ARS system. When a customer connects to the ARS server by wire or wireless communication, the ARS server guides the service menu by number and the customer wants to know. If you enter the corresponding number on the wired or wireless terminal, the ARS server will retrieve the information to inform the customer.

At this time, the voice recognition system requests a menu of a desired service by voice, not by pressing a button on the terminal.

However, the voice recognition rate is often lowered due to a lot of words that need to be recognized in the surrounding noise or vibration or the voice recognition system, or when there are many similar words among all the words to be recognized. There are many limitations to relying on.

Therefore, in recent years, an information providing model in which a counseling knowledge is electronically documented and a counselor searches a related document through a search and provides information requested by a customer has been widely used.

In this case, a search method by keyword input for searching a corresponding document by keyword input for necessary information is common.

However, the keyword search method is difficult to review many documents during a time-consuming phone call when the keyword is used in many documents. If the keyword entered is not listed in the knowledge document, the search results do not appear as a search result. There is a problem with noise in the results and the related documents are not properly retrieved.

In particular, the electronically documented information about the information that the customer wants to know is mainly the terminology used in the call center. In fact, the customer uses non-technical and everyday terminology so that the counselor can immediately see the difference between the terms used by both parties. If it is not recognized as a problem, it is difficult to retrieve related information.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a call center counseling method and counseling system using voice recognition and tagging that can reduce the burden of the counselor's work and improve the work environment of the counselor.

In addition, the present invention uses not only the terminology used by the counselor using voice recognition and tagging technology, but also the non-professional and everyday terminology that the customer understands and uses as a keyword to search for related knowledge, thereby being friendly to the everyday language used by the customer. An object of the present invention is to provide a call center counseling method and counseling system using voice recognition and tagging that can provide information through person search.

In addition, the present invention is to provide a call center counseling method and counseling system using voice recognition and tagging that can significantly reduce the search time of the information requested by the customer in the telephone response of the call center struggling time to realize customer satisfaction The purpose.

In order to achieve the above object, the present invention provides a call center counseling method for a request for providing information from a customer through a wired or wireless terminal, the keyword input step of inputting a keyword extracted from the conversation contents of the customer and the counselor to the voice recognition unit; The voice recognition unit recognizes the keyword existing in the conversation contents of the counselor and the tagging step, and generates tagging information by tagging the related document and the keyword searched by the counselor according to the conversation contents and the conversation contents of the client and the counselor. The voice recognition unit recognizes the existing keyword, and provides a call center counseling method using voice recognition and tagging, comprising the step of providing an information based on the tagging information for the keyword.

The keyword input step may further include recording the conversation contents of the customer and the counselor, extracting a plurality of keywords from the recording contents, and inputting the extracted plurality of keywords to a voice recognition unit. It is preferable.

In addition, the keyword input step is preferably updated periodically.

The tagging step may further include storing tagging information generated by tagging the searched related document and the keyword in a tagging DB. The tagging step may include: grouping the tagging information into a plurality of categories; Indexing The method may further include storing information grouped and indexed into the plurality of categories in an indexing DB.

Furthermore, the tagging step may further include analyzing and tagging the tagging information and storing the analysis and statistical information in a statistical DB.

Preferably, the voice recognition unit recognizes the keyword existing in the conversation contents of the customer and the counselor, retrieves the information about the keyword from the tagging DB, the indexing DB or the statistics DB, and retrieves the related document based on the retrieved information. It is desirable to demonstrate.

Meanwhile, in order to achieve the above object, the present invention provides a call center counseling method for a request for providing information from a customer through a wired or wireless terminal, wherein a conversation DB and a conversation stored in the recording DB are recorded and stored. The keyword extracted from the contents is input, and the tagging information is generated by tagging the voice recognition unit that recognizes the keyword existing in the conversation contents of the customer and the counselor, and the related document searched by the counselor according to the keyword and the conversation contents. The present invention provides a call center counseling system using voice recognition and tagging, comprising: a tagging unit for storing and a display unit for displaying a related document based on tagging information on keywords recognized through the voice recognition unit.

Here, the tagging unit preferably includes a tagging DB for storing the tagging information.

The method may further include a clustering and indexing unit for grouping and indexing the tagging information into a plurality of categories and an indexing DB for storing information grouped and indexed into the plurality of categories.

The apparatus may further include a statistics unit analyzing and analyzing the tagging information, and a statistics DB storing the analysis and statistical information.

In addition, it is preferable to further include a search unit for searching for information on the keyword in the tagging DB, indexing DB or statistics DB through the keywords recognized through the speech recognition unit.

According to the present invention, the counselor's work burden is eliminated without the counselor searching and acquiring or memorizing a large amount of information through printed matter, and a search process such as generating a keyword for searching related information is unnecessary. This has the effect of improving the work environment of the counselor.

In addition, in the present invention, not only the terminology used by the counselor using voice recognition and tagging technology, but also the non-professional and everyday terminology that the customer understands and uses, becomes a keyword, so that related knowledge is searched, and thus, to the everyday language used by the customer. There is an advantage that can provide information through friendly search.

In addition, in the present invention, in the telephone response of the call center struggling time, the information requested by the customer is automatically displayed through the tagging technology and the search of the information through the voice recognition and tagging, which greatly reduces the search time of the information. The call center counseling method and counseling system used can be provided.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the embodiments of the present invention described below are provided to enable those skilled in the art to more easily understand the present invention, and the scope of the present invention is not limited to the described embodiments.

1 is a block diagram schematically showing the configuration of the customer to the counselor through a wired or wireless terminal.

Referring to Figure 1, the call center counseling system using voice recognition and tagging according to the present invention, the customer has a wired or wireless terminal (1), a wired or wireless carrier's external exchange (2), the customer's request Its own exchange 5 in the call center 3 providing information on the call center, a call distribution server 7 distributing a plurality of calls from the customer, and a plurality of users providing information about the customer's request through a call with the customer. It consists of a consultant (9).

First, a customer enters a telephone number that can be connected to the call center 3 through a wired or wireless communication network to the wired or wireless terminal 1 to connect to the call center 3.

In the process of accessing the call center 3, the customer is connected to the external exchange 2 of the wired communication company or the wireless communication company through the wired or wireless terminal 1, and then the self-exchanger held by the call center 3 Connected to (5).

The external exchange 2 of the wired communication company or the wireless communication company and the own exchange 5 held by the call center 3 are connected by a dedicated line.

After connecting to the own exchange 5 of the call center 3, the call distribution server 7 is connected to the call distribution server 7 distributes a plurality of calls from the customer, the counselor to provide information to the customer ( 9) and the final connection.

The counselor 9, who is connected by phone through the above process, proceeds to consult the information requested by the customer through a call with the customer.

2 is a block diagram of a call center counseling system using voice recognition and tagging according to the present invention.

The call center counseling system using voice recognition and tagging according to the present invention includes a recording DB 10, a voice recognition unit 20, a tagging unit 30, a tagging DB 35, a clustering and indexing unit 40, and an indexing DB ( 45), the statistics section 50, the statistics DB 55, the search section 60 and the display section 70.

In the recording DB 10, when a call is connected between the customer and the counselor 9, the conversation content is recorded and stored between the customer and the counselor 9.

At this time, the conversation content may be stored in the form of a wave file or the like, and the recording DB 10 may use a conversation recording DB which is essentially operated in most call centers 3.

The voice recognition unit 20 inputs a keyword extracted from the conversation contents between the customer and the counselor 9 stored in the recording DB 10, and the keyword exists in the conversation contents between the customer and the counselor 9. If so, the keyword is recognized.

The process of extracting keywords from the conversation contents between the customer and the counselor 9 stored in the recording DB 10 may be performed manually.

The keywords may include frequently asked questions and answers by customers, frequently asked questions, vocabulary, synonyms, or synonyms for the brand name and related concepts processed in the call center.

The keyword also includes both terms used by the customer to describe the request and terms used by the counselor 9 to provide information about the request of the customer.

That is, not only the terminology used by the counselor 9 but also the non-professional and everyday terminology that the customer understands and uses may be a keyword.

For example, if a customer wants to get information about "cook TV" but actually says "KT IPTV" instead of "cook TV", the term "cook" used by the counselor 9 "TV" and the customer's "KT IPTV" can both be keywords.

When the keyword is extracted from the recording DB 10 and the extracted keyword is input to the voice recognition unit 20, a process of recognizing the keyword in a conversation between a customer and a counselor will be described. If the above keyword is present in the voice of the customer and the counselor in the process of executing and executing the conversation between the customer and the counselor 9 stored therein, only the keyword portion is extracted separately.

The extracted keyword is stored in, for example, a sample voice storage unit (not shown) of the voice recognition unit 20, and when the keyword is used in a conversation between a customer and a counselor, the voice recognition unit 20 recognizes this and recognizes the voice. The voice comparison unit (not shown) of the unit 20 compares the voice stored in the sample voice storage unit.

Through the comparison process, the voice recognition unit 20 recognizes the existence of the keyword in the conversation between the customer and the counselor.

In addition, the voice recognition unit 20 converts the plurality of keywords recognized in real time during the conversation between the customer and the counselor 9 into text for tagging.

The tagging unit 30 tags and stores the keywords recognized through the voice recognition unit 20 with related documents searched by the counselor 9 to provide information on the request information of the customer.

Tagging is mainly used on blogs or web pages, and the administrator of a site processes keywords or images of the site's image or text into related topics or categories. Keyword).

In the present invention, it is to be understood that the counselor 9 processes the searched documents in related topics or categories, and mainly the terminology used by the counselor for information that the customer wants to know or the non-professional that the customer understands and uses. Everyday terms become keywords.

The tagging unit 30 includes a tagging DB 35. The tagging DB 35 stores tagging information generated by tagging the keyword and the related document searched by the counselor 9.

That is, the tagging unit 30 combines the keyword with a unique ID of a related document and stores the tagging information in a tagging DB 35 that stores tagging information.

In this case, various types of tagging information may be stored in the tagging DB 35. For example, contents such as the schema of FIG. 3 may be stored.

Referring to FIG. 3, a document table storing document information may store a unique ID of a document, a document title, a content, a date of creation, and a create_date.

In addition, a tag-list table for storing keywords for tagging may store a tag ID and a tag, and knowledge-tags for linking the document table and the tag-list table. (knowledge-tag) A table may exist.

In addition to the above tables, various tables may be formed and added as necessary.

Referring back to FIG. 2, the clustering and indexing unit 40 groups and indexes the tagging information into a plurality of categories.

That is, the tagged documents can be grouped with related documents, which can be grouped by various categories, and indexed with respect to the grouped documents.

Here, an algorithm for grouping documents based on the tagging information is required, and various known clustering algorithms may be used.

For example, various known clustering algorithms, such as K-means algorithm, Hierarchical Clustering algorithm, Expectation Maximization algorithm, can be used.

Grouping work (clustering) as described above, in consideration of the nature of the work and the amount of work, it is preferable to work at the same time at night or do not work with the call center (3) work system at a time.

Because clustering algorithms usually require a lot of computation and consume a lot of time and resources, it is more efficient to work at a time on a regular basis.

The clustering and indexing unit 40 includes an indexing DB 45. The indexing DB 45 stores information grouped and indexed into a plurality of categories.

The statistics unit 50 provides or updates various statistical data by analyzing information stored in a tagging DB or an indexing DB.

For example, based on tagging information, the tag cloud (counting tags, counting the number of views per document, etc.) to construct a tag cloud (visually arranged in a web site based on importance or popularity by analyzing tags), The tagging vocabulary list by date may be analyzed through the statistics unit 50.

The statistics unit 50 includes a statistics DB 55, which stores information analyzed and statistically analyzed through the statistics unit 50.

The analysis and statistical information of the statistics unit 50 may have various contents according to the information analysis request of the call center (3).

The search unit 60 searches for information on the keyword in the tagging DB 35, the indexing DB 45, or the statistics DB 55 through the keywords recognized by the voice recognition unit 20.

The keyword recognized by the voice recognition unit 20 among the conversation contents between the customer and the counselor 9 becomes input information of the search unit.

The tagging DB 35, indexing DB 45 or statistics DB 55 is searched by the keyword, and documents in the tagging DB 35 and related contents in the indexing DB 45 are grouped together. Analysis and statistical information in the documents or statistics DB 55 are output in various forms to the display unit 70 to be described later.

In the present invention, not only the terminology used by the counselor 9, but also the non-professional and everyday terminology that the customer understands and uses can be tagged as the keywords and searched through the search unit 60. In addition, it is possible to provide information through a search that is friendly to everyday language used by customers.

In fact, the terms used by the counselor (9) are often different from the terms used by the client for the request for specific information of the client. Even though the term used by the client and the consultant (9) are different, the terms are different from each other. By tagging the related documents, the same related documents can be displayed on the display unit 70 to be described later, no matter what term.

The display unit 70 displays the related document searched by the counselor 9 to provide information on the request information of the customer or the information searched through the search unit 60 so that the counselor can view the information.

In this case, the information displayed on the display unit 70 may be output in various forms such as document summary contents, graphic diagrams, and two-dimensional tables in consideration of the readability of the counselor 9.

The display unit 70 may be a monitor, a terminal, or the like of the computer of the counselor 9.

As described above, a ranking algorithm may be applied between the tagged documents and the grouped documents displayed on the display unit 70.

In this case, an algorithm for applying a ranking to a tagged document or a clustered and indexed document using statistical information such as tag counts, document counts, and tagging vocabulary lists by date analyzed through the statistics unit 50 may be applied. have.

Ranked documents as described above may be stored in the statistics DB 55, and may be displayed on the display unit 70 in the order of the documents most suitable for the keywords recognized by the speech recognition unit 20.

In addition, for grouped documents, it is also necessary to keep the number of documents in the group appropriately so that there are not too many documents in a group.

This is because, if too many documents are displayed on the screen, the counselor 9 may not check all the documents in a short time, and there may be a problem in that the provision of information according to the checking of documents is delayed.

4 is a flowchart illustrating a call center counseling method using voice recognition and tagging according to the present invention.

Referring to Figure 4, via the wired terminal or wireless terminal receives a request for providing information on the basis of inquiries, failure reports, complaints, etc. for the product from the customer (S210), a call is connected between the customer and the counselor to talk Begins.

The conversation contents between the customer and the counselor are recorded (S215), and a plurality of keywords are extracted from the recorded contents (S220).

The keywords may include frequently asked questions and answers by customers, frequently asked questions, vocabulary, synonyms, or synonyms for the brand name and related concepts processed in the call center.

Here, the keyword includes both terms used by a customer to describe a request and terms used by a counselor to provide information about the request of the customer.

In other words, not only the terminology used by the counselor 9, but also the non-technical and everyday terminology that the customer understands and uses becomes a keyword.

Next, the extracted keywords are stored in the sample voice storage of the voice recognition unit 20 in order to input a plurality of keywords extracted in the voice recognition unit 20 (S225).

The keyword must be continuously reinforced through the continuous update, and the process of extracting a new keyword by continuously or periodically recording the conversation between the customer and the counselor should be repeated.

The extracted new keyword is input to the voice recognition unit 20 through continuous input work.

The data accumulated through the above process (S210 ~ S225), that is, the first step is a process that must be preceded to provide information to the customer in the future.

Having the voice recognition unit 20, the keyword is input through the first step, and a request for information provision is received from the customer (S230), when the conversation between the customer and the counselor starts, the voice recognition unit 20 operates together. Done.

The counselor listens to the customer's request and finds a relevant document that can be answered through a search (S235), and explains the appropriate information to the customer.

At the same time, the voice recognition unit 20 recognizes the keywords present in the conversation between the customer and the counselor in real time (S240).

Here, the voice recognition unit 20 converts the plurality of keywords recognized in real time into text for tagging.

Next, keywords recognized through the voice recognition unit 20 are tagged with the related documents searched by the counselor (S245), and the tagging information is stored in the tagging DB 35 (S250).

For example, when the customer inquires the rate information for the Cook Internet phone, keywords such as "Cook Internet phone" and "rate" are recognized in real time by the voice recognition unit 20 and converted into text for tagging. It is tagged and stored with related documents.

In response to the customer's inquiries, the counselor retrieves the bill detail document of Cook Internet phone and outputs it to the display unit 70 and explains the rate system for each contract. The voice recognition unit 20 recognizes a keyword such as "contract". Also tag and store related documents.

That is, by recognizing and tagging the voices of both the customer and the counselor, even if the customer and the counselor use different terms for the same information, different terms used by the customer and the counselor are tagged in the same related documents so that they can be searched later. do.

The above process (S230 ~ S250), that is, the second step is also a process that must be preceded in the future to provide information to the customer as in the first step.

5 is a flowchart illustrating another embodiment of the second step for a call center counseling method using voice recognition and tagging according to the present invention.

In FIG. 5, only characteristic parts distinguished from FIG. 4 will be extracted and described. For convenience of description, like reference numerals refer to like elements, and description thereof will be omitted.

Referring to FIG. 5, the second step may further include grouping and indexing the tagging information into a plurality of categories (S310), and storing the information grouped and indexed into the plurality of categories in an indexing DB (S315). It may include.

The second step may further include analyzing the tagging information or the indexing information and generating statistics (S320), and storing the analysis and statistical information in the statistics DB 55 (S325).

Referring back to FIG. 4, a plurality of keywords are input to the voice recognition unit 20 through the first and second steps described above, and the keywords are inputted through the voice recognition unit 20. Information such as tagged documents, indexed documents, analysis and statistical information can finally provide a more convenient consultation environment for call center consultation.

In other words, in the third step, the voice recognition unit 20 and the tagging information are used to conveniently provide information on the request for providing the information of the customer, and the method will be described below.

Through the wired terminal or the wireless terminal, a request for providing information on the basis of inquiries, trouble reports, complaints, etc. is received from the customer (S255), and a call is connected between the customer and the counselor to start a conversation.

The voice recognition unit 20 recognizes a keyword existing in the conversation between the customer and the counselor (S260), and the tagging information related to the keyword is retrieved from the tagging DB 35 through the search unit 60 ( S270).

Based on the retrieved tagging information, a related document is displayed (S275), wherein the related document is automatically displayed on the display unit 70 without a searching process such as a keyboard typing of a counselor.

That is, since the document related to the answer is automatically displayed before the counselor directly searches for the answer, a search process, such as coming up with a keyword for searching for related information, is unnecessary, so that the work environment of the counselor may be improved.

In addition, the counselor does not have to memorize all the relevant knowledge of the information requested by the customer, and can dramatically reduce the time required to search for answers to the customer's request in the time-consuming call center call. .

6 is a flowchart illustrating a call center counseling method according to a preferred embodiment of the present invention.

Through the wired terminal or the wireless terminal, a request for providing information on the basis of inquiries, trouble reports, complaints, etc., is received from the customer (S410), and a call is connected between the customer and the counselor to start a conversation.

The conversation between the customer and the counselor is recorded (S415), the counselor hears the customer's request and finds a relevant document that can be answered through a search (S420), and explains the appropriate information to the customer. .

At the same time, the voice recognition unit 20 operates together to recognize the keywords present in the conversation between the customer and the counselor in real time (S425).

Next, search for the presence or absence of tagging information related to the recognized keywords (S430), and if there is tagging information, the tagging information is extracted from the tagging DB 35 (S435), and the related document based on the tagging information is retrieved. It is displayed on the display unit 70.

Therefore, the counselor can see the relevant document displayed on the display unit 70, and can provide an appropriate answer to the customer.

If the tagging information related to the keywords recognized by the voice recognition unit 20 does not exist, the keyword is tagged with a related document searched by a counselor (S445) and the tagging information is stored in the tagging DB 35. (S450).

When the conversation between the customer and the counselor is terminated, recording is terminated (S460). If the conversation is not terminated (S455), the above steps (S425 to S455) are repeated.

When the conversation recording is finished, an input operation is performed to extract a plurality of keywords from the recorded contents (S465) and to input a plurality of keywords extracted to the voice recognition unit 20 (S470).

The above description is merely illustrative of the technical idea of the present invention, and those skilled in the art to which the present invention pertains may make various modifications and changes without departing from the essential characteristics of the present invention. Therefore, the embodiments described in the present invention are not intended to limit the technical idea of the present invention but to explain, and the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

1 is a block diagram schematically showing the configuration until the customer is connected to the counselor through a wired or wireless terminal,

2 is a block diagram of a call center counseling system using voice recognition and tagging according to the present invention;

3 is a table illustrating an example of tagging information stored in a tagging DB.

4 is a flowchart illustrating a call center counseling method using voice recognition and tagging according to the present invention.

5 is a flowchart illustrating another embodiment of a second step among the steps shown in FIG. 4 for a call center counseling method using voice recognition and tagging according to the present invention;

6 is a flowchart illustrating a call center counseling method according to a preferred embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

10: recording DB 20: voice recognition unit

30: tagging unit 35: tagging DB

40: clustering and indexing unit 45: indexing DB

50: statistics unit 55: statistics DB

60: search unit 70: display unit

Claims (12)

In a call center consultation method for requesting information provision from a customer through a wired or wireless terminal, A keyword input step of inputting a keyword extracted from the conversation contents of the customer and the counselor into the voice recognition unit; A tagging step of recognizing the keyword existing in the conversation contents of the customer and the counselor, and generating tagging information by tagging the related document and the keyword searched by the counselor according to the conversation contents; And An information providing step of recognizing the keyword existing in the conversation contents of the customer and the counselor and presenting a related document based on tagging information on the keyword; Call center counseling method using voice recognition and tagging comprising a. The method of claim 1, The keyword input step, Recording the conversation contents of the customer and the counselor; Extracting a plurality of keywords from the recording; Inputting the extracted plurality of keywords into a voice recognition unit; Call center counseling method using voice recognition and tagging, characterized in that it further comprises. The method of claim 1, The keyword input step is a call center counseling method using voice recognition and tagging, characterized in that is updated periodically. The method of claim 1, The tagging step may include storing tagging information generated by tagging the searched related document and the keyword in a tagging DB; Call center counseling method using voice recognition and tagging, characterized in that it further comprises. The method of claim 1, The tagging step may include grouping and indexing the tagging information into a plurality of categories; Storing the information grouped and indexed into the plurality of categories in an indexing DB; Call center counseling method using voice recognition and tagging, characterized in that it further comprises. The method of claim 1, The tagging step may include analyzing and tagging the tagging information; And Storing the analysis and statistical information in a statistics DB; Call center counseling method using voice recognition and tagging, characterized in that it further comprises. The method according to any one of claims 4 to 6, The information providing step, The voice recognition unit recognizes the keyword existing in the conversation between the customer and the counselor, retrieves information about the keyword from the tagging DB, the indexing DB or the statistics DB, and displays a related document based on the retrieved information. Call center counseling method using voice recognition and tagging. In a call center consultation system for requesting information provision from customers through wired and wireless terminals, A recording DB in which conversations between the customer and the counselor are recorded and stored; A voice recognition unit for inputting a keyword extracted from the conversation contents stored in the recording DB and recognizing the keyword existing in the conversation contents of the customer and the counselor; A tagging unit configured to generate tagging information by tagging related documents searched by a counselor according to the keyword and the contents of the conversation; And A display unit for displaying a related document based on tagging information of a keyword recognized through the voice recognition unit; Call center consultation system using voice recognition and tagging, comprising a. The method of claim 8, The tagging unit, a call center counseling system using voice recognition and tagging, characterized in that it comprises a tagging DB for storing the tagging information. The method of claim 8, A clustering and indexing unit for grouping and indexing the tagging information into a plurality of categories; And An indexing DB for storing information grouped and indexed into the plurality of categories; Call center counseling system using voice recognition and tagging, further comprising a. The method of claim 8, A statistics unit for analyzing and tagging the tagging information; And A statistics DB for storing the analysis and statistical information; Call center counseling system using voice recognition and tagging, further comprising a. The method according to any one of claims 9 to 11, A search unit searching for information on the keyword in the tagging DB, the indexing DB, or the statistics DB through the keyword recognized through the voice recognition unit; Call center counseling system using voice recognition and tagging, further comprising a.

Images