KR20210125399A - Interactive voice bot server and unmanned counsel system - Google Patents

Abstract

An interactive voice bot server and an unmanned consultation method using the same according to one embodiment of the present invention provide in processing a call, managing a scenario, and processing a function colloquially by using the interactive voice bot server that modularizes the function common to various types of voice bots. The function for processing the call converts a voice of a customer inputted from an exchange device into text and analyzes the converted text to generate an outgoing text, converts the outgoing text to a voice and transmits thereof to a customer, and enables recording per topic of the conversation. The function for managing the scenario enables inquiry of the interactive scenario and changes to the item, and the function for processing colloquially enables refinement, substitution, and rhetorical transformation of colloquial text. Therefore, the present invention is capable of allowing advisors or managers to easily view and modify a consultation scenario.

Description

Interactive voice bot server and unmanned consultation method using the same {INTERACTIVE VOICE BOT SERVER AND UNMANNED COUNSEL SYSTEM}

The present invention relates to an interactive voice bot server and an unattended consultation method using the same, and more particularly, to an interactive voice bot server that enables unattended consultation without an agent connection to a customer accessing a call center service and an unmanned using the same It's about counseling.

In general, many companies such as manufacturers and service providers operate customer centers such as call centers to provide customer service for products they sell or services they provide.

The call center connects customers and counselors using the voice guidance system to receive service requests from customers, and the counselors directly input the received service details into the computer and receive online services, such as selecting a worker to handle the service details proceed with the procedure.

However, this existing call center operation method has the inconvenience of having to go through many steps before a customer who calls the call center is connected to a counselor in the corresponding part, and there is a problem in that it is difficult to receive a service without a counselor.

In addition, in the case of a counselor, the customer's service reception details and worker information must be directly computerized to enable online management, which increases the cost of labor required for call center operation.

In order to overcome this limitation, a call center system using the ARS system was introduced, but there were many difficulties in proceeding with the service with the received contents. Even when the counselor directly listens to the voice recorded in the ARS system and processes the reception, there was a problem that the recorded voice had to be heard repeatedly, and the recorded voice was converted into text using STT (Speech To Text). Even in the case of checking , there was a problem in that it was necessary to check all the contents one by one, and there were many cases where the customer's colloquial or investigation was not accurately converted.

Korean Patent Registration 10-1179915

An object of the present invention is to solve the above problems, and in detail, to provide an interactive voice bot server capable of unattended consultation without a counselor by interlocking with the ARS system of a call center, and an unmanned reception method using the same.

The problems of the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

An interactive voice bot server according to an embodiment of the present invention for solving the above problems is interlocked with an exchange, IVR server, STT server and TTS server of a call center system to provide an unattended consultation service to customers, but via the exchange Thus, when a customer's consultation request is received from the IVR server, the call is directly connected to the exchange without going through the IVR server, and the customer's voice input from the exchange is converted into received text through the STT server, and the An outgoing text may be generated by analyzing the converted received text, the outgoing text may be converted into an outgoing voice through the TTS server, and the outgoing voice may be transmitted to the customer through the exchange.

In addition, the conversational voice bot server according to an embodiment of the present invention further includes a call processing unit for managing a conversation history and voice recording between the customer and the conversational voice bot server, wherein the call processing unit includes: A bot conversation history management unit configured to track the conversation history with the interactive voice bot server as a customer phone number; It may include a recording DB for storing the customer's voice file recorded in the recording unit, and an operating standard management unit for managing operating hours for each consultation item, comments for each time zone, and reference for connecting a counselor.

In addition, the recording unit of the interactive voice bot server according to an embodiment of the present invention indexes the voice corresponding to each query item so that the voice input from the exchange can be tracked and reproduced for each query item so that the voice can be tracked separately. can be saved as

In addition, the recording unit of the interactive voice bot server according to an embodiment of the present invention may store the voice corresponding to each query item as each individual file so that the voice input from the exchange can be tracked and reproduced for each query item.

In addition, the conversational voice bot server according to an embodiment of the present invention further comprises a chatbot server that analyzes the received text to generate an outgoing text and generates an output message for each scenario, wherein the chatbot server includes the received text. A preprocessor for preprocessing, a colloquial conversion unit for converting into spoken language based on the received text preprocessed by the preprocessor, a rhetorical conversion unit for converting a rhetoric based on the received text preprocessed by the preprocessor, and the colloquial transformation A database in which data that is the basis of colloquial conversion and rhetorical conversion in the part and the rhetorical conversion unit, respectively, is built in advance, and each outgoing text and scenario based on the colloquial words converted by the colloquial language conversion unit and the rhetoric converted by the rhetorical conversion unit It may include a comment output unit for generating and outputting an output comment.

In addition, the conversational voice bot server according to an embodiment of the present invention is a scenario management unit that provides a scenario management screen that can check and change the conversation scenario between the customer and the private conversational voice bot server in real time to a counselor terminal connected through a network The scenario management unit may further include, on the scenario management screen, a screen for inquiring and changing a query item, a data type for each query item, a question message, the number of materials, a question order, and whether an answer is required.

In addition, the unmanned consultation method using the interactive voice bot server linked to the switchboard, the IVR server, the STT server and the TTS server of the call center system according to an embodiment of the present invention receives a customer's service request from the IVR server via the switchboard. receiving, directly connecting a call between the exchange and the unmanned receptionist bot server without going through the IVR server, and sending a query message to the customer through the exchange; converting the customer's voice into an incoming text through the STT server, generating an outgoing text by analyzing the converted received text, and converting the outgoing text into an outgoing voice through the TTS server; transmitting the converted outgoing voice to the customer through the exchange; tracing the customer's voice input from the exchange for each query item, recording and storing in a reproducible form; the inquiry message and the input text; and calculating and storing the consultation details based on the recorded voice information of the customer.

In addition, in the unmanned consultation method according to an embodiment of the present invention, the step of tracing the customer's voice input from the exchange for each query item, recording and storing in a reproducible form, the customer's voice corresponds to each query item It may include the step of indexing the voice to be tracked separately and storing it as a file.

In addition, in the unmanned consultation method according to an embodiment of the present invention, the step of tracing the customer's voice input from the exchange by item, recording and storing in a reproducible form, includes the customer's voice corresponding to each query item. It may include saving each as a separate file.

In addition, the step of generating the outgoing text by analyzing the converted text of the unattended counseling method according to an embodiment of the present invention includes pre-processing the received text, colloquial conversion and rhetorical conversion based on the pre-processed received text and generating an outgoing text and an output comment for each scenario based on the converted colloquialism and rhetoric.

The non-interactive voice bot server and the unmanned reception method using the same according to an embodiment of the present invention can interwork with the ARS system of a call center to conduct unattended consultation without a counselor.

In addition, the interactive voice bot server and the unmanned reception method using the same according to an embodiment of the present invention provide a call center service by directly connecting a call to an exchange without going through an IVR server when a customer service request is received and performing a bot conversation service. By minimizing customization of the system, consultation services can be provided at low cost in a short period of time.

In addition, the interactive voice bot server and the unmanned reception method using the same according to an embodiment of the present invention modularize and provide functions common to various types of voice bots, thereby efficiently constructing a system including various types of voice bots. make it possible

In addition, the interactive voice bot server and the unmanned reception method using the same according to an embodiment of the present invention track the customer's voice for each query item and record it in a reproducible form so that a counselor or manager can easily check the customer's voice for each query item In this way, you can check and correct the received details.

In addition, the interactive voice bot server and the unmanned reception method using the same according to an embodiment of the present invention provide a scenario management screen that can check and change the conversation scenario in real time, so that the counselor or manager can easily check and modify the consultation scenario. can do.

The effect according to the present invention is not limited by the contents exemplified above, and more various effects are included in the present specification.

1 is a schematic configuration diagram of a call center system to which an unmanned receptionist bot server and an interactive voice bot server are applied according to an embodiment of the present invention.
2 is a schematic configuration diagram of a call center system for explaining the detailed configuration and operation of an unmanned receptionist bot server and an interactive voice bot according to an embodiment of the present invention.
3 is a schematic configuration diagram of a call center system for explaining the detailed configuration and operation of an unmanned receptionist bot server and an interactive voice bot according to another embodiment of the present invention.
4 is a schematic flowchart illustrating a method for the unmanned receptionist bot server according to the embodiments of FIGS. 2 and 3 to perform unattended reception.
5 is a schematic flowchart illustrating a text processing method of the interactive voice bot in FIG. 4 .
6 is an example of reception list information and correction screen provided by the unmanned receptionist bot server according to embodiments of the present invention.
7 is an example of a scenario management screen provided by an unmanned receptionist bot server or an interactive voice bot server according to embodiments of the present invention.
8 is an example of a real-time monitoring screen provided by an unmanned receptionist bot server or an interactive voice bot server according to embodiments of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the art to which the present invention pertains It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

Hereinafter, an interactive voice bot server and an unmanned consultation method using the same according to embodiments of the present invention will be described with reference to the drawings.

1 is a schematic configuration diagram of a call center system to which an unmanned receptionist bot server and an interactive voice bot server are applied according to an embodiment of the present invention.

Referring to FIG. 1 , it can be seen that each call center system is connected to a data network (network network) and a telephone network (voice network). The solid line indicates that it is connected to the data network, and the dotted line indicates that it is connected to the telephone network.

When a customer makes a call to the call center, it is connected to an exchange (PBX, 10) through the telephone network, and the exchange (10) branches to an automatic answering system (IVR, 20 or 30), and the assigned automatic answering system is an unmanned receptionist bot ( 100 or 200).

The automatic response system is called ARS (Audio Response System) or IVR (Interactive Voice Response), and in this specification, it will be described as IVR for convenience. In addition, in the present invention, a bot that receives an AS service application, delivery order, civil complaint application, etc. unmanned in the present invention is illustrated and described as unmanned reception bots 100 and 200, but this is only an embodiment.

That is, the unmanned receptionist bots 100 and 200 are only an example of an unmanned consultation bot that consults with customers unattended, and bots with other names that perform different functions depending on the content or type of consultation are possible, and the present invention covers this.

In FIG. 1, two IVRs 20 and 30, and two unmanned reception bots 100 and 200 are illustrated, but this is only an example, and may be installed in different numbers depending on the required system capacity. can

The switchboard 10 and the IVR 20 are connected to each other through a telephone network, and also to the counselor groups 3 and 5 through the telephone network. Here, the counselor groups 3 and 5 may refer to groups classified according to the contents of counseling or other criteria.

The switchboard 10 serves to branch to the IVR 20 when a request for service reception is received from the customer, and the IVR 20 allows the customer to select the desired service through a pre-built automatic response guidance message. do.

For example, when the unmanned reception bot 100 is an unmanned reception bot serving to receive an AS service reservation, and the unmanned reception bot 200 is an unmanned reception bot that serves to receive civil complaints, the IVR 20 is When a customer requests an AS service reservation, it may branch to the unmanned reception bot 100 , and when a customer requests to receive a civil complaint, it may branch to the unmanned reception bot 200 .

As the configuration of the call center system, the computer phone integrated system (CTI, 40), the recording device (50), the electronic display board (60), the web application server (WAS, 70) and the database management system (DBMS, 80) are built through the internal data network. may have been

The CTI 40 serves to efficiently distribute and manage calls coming to the call center system by combining a computer and a phone. In the CTI 40, information such as reception details, which will be described later, is displayed in real time through a computer screen during a call with a customer to enable efficient call center operation.

The recording device 50 serves to record and store the voice calls of the counselor and the customer of the call center.

The billboard 60 serves to output a screen for monitoring the operating state of the entire call center system.

The WAS 70 is a software management server that manages each application executed in the data network of the call center system, and serves to operate the application at the server end.

The DBMS 8 serves to manage a database constructed through software for information such as consultation details of customers and agents of the call center system.

The unmanned receptionist bots 100 and 200 according to an embodiment of the present invention are disposed in the “DMZ Zone” via the internal firewall 7 in the internal data network of the call center system. Here, “DMZ Zone” means an area located in the internal data network but accessible from the external data network.

The unmanned reception bots 100 and 200 are connected to an external data network via an external firewall 9 and operate as a cloud-based service interactive voice bot server 300, STT server 400, TTS server 500 and It may be connected to the DBMS 600 .

The interactive voice bot server 300 directly connects the call with the PBX 10 in response to a request when the unmanned reception bot 100 receives a request for service, etc. plays a management role. Detailed functions of the interactive voice bot server 300 will be described later.

The STT server 400 serves to convert voice data into text data, and the TTS server 500 serves to convert text data into voice data. Specifically, when voice data is received from a customer, the STT server 400 converts it into text data and transmits it to the interactive voice bot 300. When the interactive voice bot 300 generates text data to be transmitted to the customer, The TTS server 500 receives it, converts it into voice data, and delivers it to the customer through the exchange 10 .

In this specification, the unmanned reception bots 100 and 200 and the interactive voice bot 300 are configured in the form of a server, and terms can be used as an unmanned reception bot server and an interactive voice bot server, which should be interpreted as the same terms. do.

2 is a schematic configuration diagram of a call center system for explaining the detailed configuration and operation of an unmanned receptionist bot server and an interactive voice bot according to an embodiment of the present invention.

FIG. 2 shows that the unmanned reception bot 100 exists in the DMZ Zone as in FIG. 1 , and the interactive voice bot 300 , the STT server 400 and the TTS server 500 are connected to an external data network, that is, a cloud service area. The case where it exists is shown.

In addition, in FIG. 2, the STT server 400 and the TTS server 500 transmit and receive data with the unmanned reception bot 100, and the unmanned reception bot 100 and the interactive voice bot 300 transmit and receive data. However, this is only an example, and the STT server 400 and the TTS server 500 may directly transmit/receive data to and from the interactive voice bot 300 via or without the unmanned reception bot 100 , and the interactive voice bot 300 may also transmit/receive data with the exchange 10 and the IVR server 20 via or without the unmanned reception bot 100 .

When a call is received from the customer terminal 2, the exchange 10 branches to the IVR server 20 so that the customer can select a desired service. The IVR server 20 branches to the corresponding unmanned reception bot 100 when a service desired by the customer is selected, and the unmanned reception bot 100 calls the interactive voice bot 300 without passing through the IVR server 20 . A call is connected between the switchboard 10 and the interactive voice bot 300 .

In this way, directly forming a call between the exchange 10 and the interactive voice bot 300 without passing through the IVR server 20 minimizes the customization of the IVR server 20 of the existing call center system, thereby reducing costs and promptly receiving unmanned reception. Because it has the advantage of being able to build a system.

When the IVR server 20 directly requests to connect the counselor, or when it is determined that the counselor connection is necessary during the unattended reception process, a call is formed between the customer terminal 2 and the counselor terminal 4 .

The unmanned receptionist bot 100 includes a reception information processing unit 110 , a reception information storage unit 120 , an image processing unit 130 , a real-time monitoring unit 140 , and a reception information output and correction unit 150 . do.

When a customer service request is received from the IVR server 20 via the exchange 10, the reception information processing unit 110 directly connects the call to the exchange 10 without going through the IVR server 20, and from the exchange Converts the input customer's voice into received text through the STT server 400, generates an outgoing text by analyzing the converted received text, converts the outgoing test into an outgoing voice through the TTS server 500, and converts By transmitting the outgoing voice to the exchange 10, the outgoing voice is transmitted to the customer, and based on this, it serves to calculate the customer's service reception details.

The reception information storage unit 120 serves to store the service reception details calculated by the reception information processing unit 110 in the form of a database.

The image processing unit 130 transmits an image data request message of an image or video to the customer terminal 2 through the network according to the request of the reception information processing unit 110 , and receives the image data received from the customer terminal 2 . It serves to transmit to the information processing unit (110).

That is, the reception information processing unit 110 may cause the image processing unit 130 to transmit a request message for image data to the customer terminal 2 according to a preset algorithm or when there is an individual request from a counselor or manager connected to the network. have. In this case, the image data request message may be transmitted through a text message or a message through a pre-installed application.

For example, when a request for video data is transmitted in the form of a text message, the message includes a URL link, so that when a customer clicks on the link, an interface can be provided to immediately take a picture or video and upload it.

When a photo or video is received in this way, the reception information processing unit 110 may store the reception details in the reception information storage unit 120 by linking with reception details such as customer information and query items.

The real-time monitoring unit 140 serves to visualize and output the service reception status for each customer received by the reception information processing unit 110 to the counselor terminal connected by calling the network. In addition, the real-time monitoring unit 140 may provide a blocking comment input window for receiving a blocking comment to be transmitted to a specific customer on the visualized and output screen. In this case, the “blocking comment” is a message in which a counselor or manager monitoring the unmanned reception situation directly intervenes to the customer, and is converted into voice through the TTS 500 server and transmitted to the customer or may be transmitted in the form of a text message.

The reception information output and correction unit 150 serves to visualize and output the service reception details stored in the reception information storage unit 120 to the counselor terminal connected through the network. Specifically, the reception information output and correction unit 150 outputs a list of service reception details for each customer in real time, and when there is a request to search for details of reception details of a specific customer from the counselor terminal, a query item, a customer corresponding to each query item It is possible to provide a detail screen including a link icon for checking the received text of and the voice recording history for each query item corresponding to the received text of the customer.

The interactive voice bot 300 is configured by modularizing functions common to various types of unmanned counseling bots, that is, voice bots, and performs call processing, natural language processing (colloquial processing and investigation conversion, etc.) and scenario management functions. By providing these common functions in a modular way, they can be applied in common to various types of voice bots, so that when building a call center system to which various types of counseling bots are applied, customization can be minimized and system construction can be streamlined.

The interactive voice bot 300 includes a call processing unit 320 , a chatbot server 340 , and a scenario management unit 360 .

The call processing unit 320 serves to manage the conversation history and voice recording between the unmanned receptionist bot 100 and the customer, and the bot conversation history management unit 332 , the recording unit 324 , the recording DB and the operation standard management unit 328 ) is included.

The bot conversation history management unit 332 serves to manage the conversation history between the customer and the unmanned reception bot 100 to be tracked based on the customer phone number. At this time, although described as a conversation history between the customer and the unmanned reception bot 100 , the unmanned reception bot 100 calls the interactive voice bot 300 to establish a bot conversation, so the interaction between the customer and the interactive voice bot 300 is established. It could be explained by conversation history.

The recording unit 324 plays a role of tracing the voice of the customer input from the exchange 10 for each query item and recording it in a reproducible form.

The recording DB 326 serves to store the customer's voice file recorded by the recording unit 324 . The recording unit 324 also stores while recording, but the recording DB 326, which is a configuration for storing the recorded data in the form of a database, may be separately configured in this way. However, the recording unit 324 and the recording DB 326 may be arranged as one configuration.

At this time, the recording unit 324 records in a traceable format for each query item, but stores it as one file and indexes the answer section corresponding to each question item so that the response voice for each question item can be traced later or each query A method of saving the response voice for each item as each individual file can be used.

The operation standard management unit 328 serves to manage the operating hours of the unmanned receptionist bot 100, the comments for each time period, and the agent connection standards. In this case, although only the unmanned receptionist bot 100 is shown in FIG. 2 , there may be several unmanned receptionist bots 100 as in FIG. 1 . The operation standard management unit 328 can manage the operating hours of the plurality of various unmanned reception bots or unmanned consultation bots, time-specific comments, and counselor connection standards by classifying them.

In addition, it may be possible to manage operating hours, comments by time zone, and agent connection standards by classifying them based on the counseling topic rather than the unmanned receptionist bot.

The chatbot server 340 analyzes the received text to generate an outgoing text, and serves to generate output comments for each scenario, and includes a preprocessor 341, a colloquial conversion unit 342, a rhetorical conversion unit 344, and a database. It is configured to include a 346 and a comment output unit (348).

The pre-processing unit 341 serves to divide the received text into units having a meaning by performing word division, etc., of the received text.

The colloquial conversion unit 342 serves to refine or replace the received text preprocessed by the preprocessor 341 .

The rhetorical conversion unit 344 serves to stagnate or replace the rhetoric among the received text pre-processed by the pre-processing unit 341 .

The database 346 serves to pre-build and store data that is the basis of the colloquial transformation and the rhetorical transformation in the colloquial language transformation unit 342 and the rhetorical transformation unit 344, respectively.

The comment output unit 348 serves to generate and output outgoing text and output comments for each scenario based on the rhetoric converted by the colloquial and rhetorical conversion unit 344 converted by the colloquial speech conversion unit 342 .

For example, the colloquialism conversion unit 342 may manage words to distinguish “positive/negative” and use them in a bot conversation scenario. The database 346 stores “like, ok, right” as an affirmative expression, and stores “no, not, no”, etc. as a negative expression, and then the customer's voice is converted into input text through the STT server 440 . When it is received and compared with the content stored in the database 346, it may be converted in the colloquial language conversion unit 342 in such a way that it is replaced with a similar word or the phrase is refined by an algorithm built in advance. The converted text data may be stored in a text format in the reception details information.

Based on the text information converted by the colloquial speech conversion unit 342 , the message output unit 348 may identify an agent connection request intent, classify abusive language, or classify greetings, and may use it for conversation scenario management.

For example, based on the data stored in the database 346, if it is determined that the customer is swearing, a warning message may be output, and if the preset criterion is exceeded, the scenario of terminating the reception guide may be developed.

In addition, the message output unit 348 builds an algorithm in advance that determines the degree of intention of the connection request to the agent is high/medium/low based on the input text, and when it is determined that the degree of intent exceeds a certain standard, it automatically sends the message to the agent. You can develop a connecting scenario.

In addition, the rhetorical conversion unit 344 may be constructed to select a different algorithm according to the voice recognition tool applied to the call center system because the rhetorical conversion method for each voice recognition tool is different.

For example, in the case of “conversion of money”, the result of the voice recognition tool may be converted into “3,600,000 won” or “3.6 million won”, which can be set to be output as “3,600,000” or the like.

Also, in the case of “Date Conversion”, if the results of the voice recognition tool are input as “This Friday”, “Well, the day after tomorrow”, “I like the day after tomorrow”, “Twenty-seventh of March”, and “Twenty-seventh of this month”, it is set as “2020- It can be set to output as 03-27”. In particular, by applying an algorithm that automatically calculates the time based on the date the bot conversation is in progress, even when a specific date is not uttered, such as “This Friday”, accurate date conversion can be made possible even in various types of utterance data. .

In addition, in the case of “time conversion”, various cases such as the case where the input text is “10:30”, “10:30”, “noon”, and “10 minutes ago” can be recognized based on the database, and the output data is also It can be set to output by selecting from various formats.

Similarly, in the case of “Number Conversion”, if the result of the input recognition tool is “1,223”, “1000 200 20 3”, etc. 3”, etc. can be selected and set to be output.

Of course, as described above, it can be constructed so that it can be selected and used according to the voice recognition tool of the call center system, but it may be constructed so that it can be automatically selected and converted by a suitable voice recognition tool.

The scenario management unit 360 serves to provide a scenario management screen that can check and change the conversation scenario between the customer and the unmanned receptionist bot 100 in real time to the counselor terminal 4 connected through the network. The scenario management unit 360 may include, on the scenario management screen, a screen for inquiring and changing the query item, the data type for each query item, the question message, the number of materials, the order of the question, and whether an answer is required, and may provide the screen. In FIG. 2 , the counselor terminal 4 is connected to the switchboard 10 and the IVR server 20 , but this does not mean the counselor terminal 4 in the form of a telephone connected through the telephone network, but in the form of a CTI connected through the data network. I'd say it means a terminal. In FIG. 2 , for convenience, the CTI type counselor terminal 4 is illustrated as being connected to the exchange 10 and the IVR server 20 .

3 is a schematic configuration diagram of a call center system for explaining the detailed configuration and operation of an unmanned receptionist bot server and an interactive voice bot according to another embodiment of the present invention.

Another embodiment of FIG. 3 is different from the embodiment of FIG. 2 in that the interactive voice bot 300 is included as a sub-configuration of the unmanned reception bot 100, except for this, reference numerals and other configurations Since they are all the same, only these differences will be described and detailed descriptions of the rest will be omitted.

Since the unmanned receptionist bot server 100 according to another embodiment of FIG. 3 includes the interactive voice bot 300 as an internal configuration, the interactive voice bot 300 is not placed in an external data network in the form of a cloud service and is DMZ Place in Zone.

Of course, even in this case, the STT server 400 and the TTS server 500 may be disposed in an external data network.

In this way, when the interactive voice bot 300 is disposed in the internal data network, it may be possible to construct an unmanned reception system with a faster speed.

In addition, although FIG. 3 shows that the interactive voice bot 300 is included in the unmanned reception bot 100 , the interactive voice bot 300 is arranged in the DMZ Zone separately from the unmanned reception bot 100 . will also be possible

4 is a schematic flowchart illustrating a method for the unmanned receptionist bot server according to the embodiments of FIGS. 2 and 3 to perform unattended reception, and FIG. 5 is a schematic flowchart illustrating a text processing method of the interactive voice bot in FIG. .

When there is a request for reception, such as service, from the customer terminal 2, when a call is set between the exchange 10 and the unmanned reception bot 100 or the interactive voice bot 300, an unmanned reception initial guidance message is sent to the customer through the TTS server 500 It is transmitted as a voice to (S410).

Then, a query message for the unmanned reception item is transmitted by voice to the customer through the TTS server 500 according to the following pre-built scenario (S420).

When the customer's answer voice information is received (S430), it is converted into text through the STT server 400 (S440), and the conversational voice bot is called (S450) to input the converted text.

Referring to FIG. 5 , the called conversational voice bot 300 performs text preprocessing (S510), calls an algorithm corresponding to the query item (S520), performs colloquial conversion and rhetorical conversion to appropriately convert text. (S530)

Referring back to FIG. 4 , the interactive voice bot 300 determines whether a preset answer criterion is suitable based on the results of the text conversion after the text preprocessing, the colloquial conversion, and the rhetorical conversion above ( S460 ).

If it is determined that the answer is correct, the process returns to step S420 and continues to inquire about the next item.

If it is determined that the answer is inaccurate, it is determined whether a preset operation standard has been exceeded ( S470 ). In this case, the preset operation criterion may be set based on the number of times the answer criterion is inappropriate or the level of the converted text data.

If it is determined that the operating standard has been exceeded, the reception failure process is processed (S475) and the unmanned reception dialog is terminated. In this case, the acceptance failure processing is performed only when the question is completed for all the query items, and when the query items remain, the query can be continued for the remaining query items.

If it is determined that the operating standards are satisfied, the process returns to step S420 and the same item is reviewed.

When all answers to the pre-set query items are received through this process, an unmanned reception completion message is transmitted (S480), and the reception information is stored in a DB form (S490).

6 is an example of reception list information and correction screen provided by the unmanned receptionist bot server according to embodiments of the present invention.

Referring to FIG. 6 , the unmanned receptionist bot 100 according to the present invention may check reception list information and a correction screen displayed on a counselor's terminal (eg, CTI, etc.) connected through a network.

On the basic screen, a list of receptions received by each customer for a certain period is displayed, and the reception list information includes reception number, product group, model, date of visit request, phone number, address, name of a consultant, and processing status.

The basic screen of FIG. 6 shows the AS service reception screen for a specific product, and the processing status is divided into “incomplete”, “pending” and “completed” according to the pre-built algorithm of the unmanned receptionist bot 100 and displayed. have. For example, “incomplete” may be a case in which some of the query items have failed to receive, and “hold” is a pending processing according to a customer interruption or inaccurate status determination according to a preset algorithm of the unmanned reception bot 100 can be “Complete” may be displayed when the unmanned receptionist bot 100 satisfies a preset criterion.

Regardless of the processing status, if the counselor selects and double-clicks a specific reception item, a pop-up window at the bottom right of the detailed reception history screen may be presented. In the pop-up window, detailed reception details of the case are displayed, each question item is displayed, and text information converted into colloquial and rhetorical by the interactive voice bot 300 is displayed on the right side of each question item, and the corresponding text information is displayed on the right side of each text information A link icon is disposed so that the customer's original voice corresponding to the text information can be checked.

If the counselor clicks the icon on the right, the original voice corresponding to the question item can be checked. Therefore, the counselor can check the details of the incomplete case and check the voice of the missing item, immediately supplement the item, save it, and process the completion. Of course, you can also check completed or pending cases and make corrections.

In this case, the counselor can quickly and easily check and modify the received information without having to check all the consultation details.

In addition, such a screen degree may be displayed automatically in conjunction with a phone call between the customer and the counselor.

7 is an example of a scenario management screen provided by an unmanned receptionist bot server or an interactive voice bot server according to embodiments of the present invention.

Referring to FIG. 7 , it is possible to check the scenario management screen displayed on the terminal of the counselor connected through the network. It shows the application scenario for after-sales service for products, and the counselor or manager displays a screen where the name of the inquiry item, the data type for the inquiry item, the inquiry message, the number of materials, the order of the inquiry and whether the answer is required or not can be entered or selected.

The reason that “data type” can be selected is to select a data type suitable for the properties of the query item so that the text processing algorithm for the corresponding query item can be selected. For example, in the case of "address" and "date", more accurate text analysis and conversion can be performed by a pre-built database and algorithm.

The number of materials means the number of times to do the material in response failure processing, the display sequence number (question sequence number) means the order to be asked, and the required answer indicates information for reference in response failure handling when an answer is not answered.

8 is an example of a real-time monitoring screen provided by an unmanned receptionist bot server or an interactive voice bot server according to embodiments of the present invention.

Referring to FIG. 8 , a real-time monitoring screen displayed on a terminal of a counselor or manager connected through a network may be checked.

On the left, a list of unattended receptions currently in progress is displayed along with a phone number and a query item currently in progress.

And on the far right, real-time analysis information of the conversation is displayed.

The manager can monitor the real-time unattended reception status through such a screen, and when the manager decides that an agent connection is necessary, he can input a blocking comment (manager's comment) and send it to the customer.

As shown in FIG. 8 , when a text is entered in the manager's comment field and a transmission icon is clicked, the corresponding text is converted into voice through the TTS server 500 and transmitted to the customer or may be transmitted to the customer's terminal in the form of a text message.

As described above, the unmanned receptionist bot server and the unmanned reception method using the same according to embodiments of the present invention have been described.

Those of ordinary skill in the art to which the present invention pertains will understand that the present invention may be embodied in other specific forms without changing the technical spirit or essential features thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

2: Customer terminal 3: Agent group
4: agent terminal 5: agent terminal
7: Internal firewall 9: External firewall
10: Exchange (PBX) 20: Automatic answering system (IVR)
30: Automatic Answering System (IVR) 40: Computer Telephone Integration System (CTI)
50: recording equipment 60: electronic sign board
70: Web Application Server (WAS) 80: Database Management System (DBMS)
100: unmanned reception bot 110: reception information processing unit
120: reception information storage unit 130: image processing unit
140: real-time monitoring unit 150: reception information output and correction unit
200: unmanned receptionist bot 300: interactive voice bot
320: call processing unit 322: bot conversation history management unit
324: recording unit 326: recording DB
328: operation standard management unit 340: chatbot server
341: preprocessor 342: colloquial conversion unit
344: investigative transformation unit 346: database
348: comment output unit 360: scenario management unit
400: STT server 500: TTS server
600: Database Management System (DBMS)

Claims (8)

In the interactive voice bot server that provides unattended consultation service to customers by interworking with the switchboard, IVR server, STT server and TTS server of the call center system,
The interactive voice bot server,
When a customer's consultation request is received from the IVR server via the exchange, a call is connected to the exchange, the customer's voice input from the exchange is converted into received text through the STT server, and the converted received text is configured to generate an outgoing text by analyzing the outgoing text, convert the outgoing text into an outgoing voice through the TTS server, and transmit the outgoing voice to the customer through the exchange,
a call processing unit for managing a conversation history and voice recording between the customer and the interactive voice bot server; and
and a chatbot server that analyzes the received text to generate outgoing text, and generates output comments for each scenario,
The call processing unit,
a bot conversation history management unit configured to track the conversation history between the customer and the conversational voice bot server with a customer phone number;
a recording unit that tracks the customer's voice input from the exchange for each query item and records it in a reproducible form;
a recording DB for storing the customer's voice file recorded in the recording unit; and
Includes an operation standard management unit that manages consultation operation hours for each consultation item, comments by time period, and agent connection standards,
The chatbot server is
a preprocessor for preprocessing the received text;
a colloquial conversion unit for converting the received text pre-processed in the pre-processing unit into colloquial speech;
a rhetorical conversion unit for converting a rhetoric based on the received text preprocessed by the preprocessor;
a database in which data that is a basis for colloquial conversion and rhetorical conversion in the colloquial conversion unit and the rhetorical conversion unit, respectively, is built in advance; and
and a ment output unit for generating and outputting outgoing texts and output comments for each scenario based on the rhetoric converted by the colloquial speech conversion unit and the rhetorical conversion unit,
The conversational speech bot server, characterized in that the colloquial speech conversion unit and the rhetorical conversion unit perform colloquial conversion and rhetorical conversion by applying a preset text processing algorithm corresponding to the data type for each query item. According to claim 1,
The audio recording unit is an interactive voice bot server that indexes and stores the voices corresponding to each query item as a file so that the voice corresponding to each query item can be tracked and reproduced for each query item by tracking the voice input from the exchange. According to claim 1,
The audio recording unit is an interactive voice bot server that stores the voice corresponding to each query item as each individual file so that the voice input from the exchange can be tracked and reproduced for each query item. According to claim 1,
Further comprising a scenario management unit that provides a scenario management screen for checking and changing the conversation scenario between the customer and the interactive voice bot server in real time to the counselor terminal connected through the network,
The scenario management unit is an interactive voice bot server including a screen for inquiring and changing the query item, the data type for each query item, the query message, the number of materials, the order of the question, and whether an answer is required on the scenario management screen. In the unmanned consultation method using an interactive voice bot server linked to a call center system switchboard, IVR server, STT server and TTS server,
Receiving the customer's service request reception from the IVR server via the exchange;
directly connecting a call between the exchange and the unmanned receptionist bot server;
sending a query message to the customer through the exchange;
converting the customer's voice input from the exchange into received text through the STT server;
generating an outgoing text by analyzing the converted received text;
converting the outgoing text into an outgoing voice through the TTS server;
transmitting the converted outgoing voice to the customer through the exchange;
tracing the customer's voice input from the exchange for each query item, recording and storing in a reproducible form;
Calculating and storing service reception details based on the query message, the received text, and the recorded customer voice information,
The step of analyzing the converted received text to generate an outgoing text includes:
pre-processing the received text;
performing colloquial conversion and rhetorical conversion based on the pre-processed received text; and
and generating an output message for each outgoing text and scenario based on the converted colloquialism and rhetoric,
The step of performing the colloquial transformation and rhetorical transformation includes performing the colloquial transformation and rhetorical transformation by applying a preset text processing algorithm corresponding to the data type for each query item. 6. The method of claim 5,
The step of tracing the voice of the customer input from the exchange for each query item, recording and storing in a playable form,
and indexing the customer's voice so that the voice corresponding to each query item is separately tracked and stored as a file. 6. The method of claim 5,
The step of tracing the voice of the customer input from the exchange for each query item, recording and storing in a playable form,
and storing the customer's voice corresponding to each query item as each individual file. 6. The method of claim 5,
The method further comprising the step of providing a scenario management screen for checking and changing the conversation scenario between the customer and the interactive voice bot server in real time to the counselor terminal connected through a network,
The scenario management screen includes a screen for inquiring and changing a query item, a data type for each query item, a question message, the number of materials, a question order, and whether an answer is required.

Images