TW201901660A - Voice number-taking method and system to regenerate and/or display the number-taking and reply message - Google Patents

Abstract

In a voice number-taking method and system, a servo unit recognizes a voice inquiry message obtained from an user located at an user end of a target service mechanism related to the desired service item by utilizing a voice recognition technology. Semantic analysis technology is used for retrieving all the questions and answers stored in a service database built in the servo unit and generating a target answer data having the maximum degree of matching with the voice inquiry messages, which is not lower than a preset threshold value. The system transmits a number-taking reply message containing the target answer data and a service serial number of the target service mechanism to the user terminal through a communication network, so that the user terminal can regenerate and/or display the number-taking and reply message.

Description

Voice taking method and system

The invention relates to a method and system for taking a number, in particular to a method and system for taking a voice.

The existing bank picking method usually has to operate a user interface provided by a banking machine or a customer itself, and the numbering opportunity prints corresponding to the user interface according to the input generated by the user interface. A list of bank service suits. The bank counter can serve the customer in the order of the serial numbers displayed on all the number sheets printed by the numbering machine.

However, it is necessary for the bank to take the method of manual operation, and if the user interface of the numbering machine is not correctly operated according to the desired banking service, it may result in an incorrect number. Therefore, there is still much room for improvement in the existing bank picking method.

Accordingly, it is an object of the present invention to provide a method of voice taking that overcomes the shortcomings of the prior art.

Therefore, the present invention provides a voice numbering method, which is implemented by a voice numbering system, which includes a user terminal located at a target service organization, and a connection via a first communication network. The server unit of the client includes the following steps: (A) at least one service database is constructed by the server unit, and the service database stores a plurality of problems corresponding to a plurality of different service items respectively. Answer data, each question and answer data includes a question data portion related to one of the service items of the service item, and an answer data portion indicating the corresponding service item; (B) by the user end, Receiving a request for a voice message containing at least the voice query message to the server unit via the first communication network when receiving a voice inquiry message from a user and related to a desired service item; (C) The serving unit converts the voice query message into an original text by using a voice recognition technology when receiving the request for the number from the user end Content, and analyzing the original text content by semantic analysis technology to obtain an analysis text content, and determining whether there is a target problem data portion in all problem data portions stored in the service database, and having a content with the analysis text content (D) by the servo unit, when the server unit determines that there is a target problem data part in all the problem data portions stored in the service database, the service database is stored. One of the answer data sections corresponds to the answer data part of the target question data section as a target answer data part, and at least one service is generated according to at least the target answer data part and the current service status of the target service organization. The serial number of the serial number replies to the message, and when it is determined that there is no target problem data portion in all the problem data portions stored in the service database, an error message indicating that the voice query message cannot be recognized is generated, and the response is received. The request to reply to the generated message or the The error message is transmitted to the user terminal via the first communication network; and (E) the user terminal regenerates the number of the reply message or the error message when receiving the message from the server unit Replying to the message or the error message so that the user can obtain the reply message or the error message on at least one of the hearing and the visual.

Accordingly, it is another object of the present invention to provide a voice taking system that overcomes the shortcomings of the prior art.

Accordingly, the present invention provides a voice taking system that includes a user terminal and a servo unit. The user terminal is located at a target service organization and includes a user interface unit including a radio module, a communication unit connected to the first communication network, and a user interface unit and the communication unit. Processing unit. The server unit is connected to the first communication network, and is configured with a service database, where the service database stores a plurality of question and answer data respectively corresponding to a plurality of different service items, and each question and answer data includes a related One of the service items corresponds to the problem data part of the service item, and an answer data part indicating the corresponding service item.

When the receiving module of the user end receives a voice inquiry message from a user and related to a desired service item, the processing unit includes, by the communication unit, at least the voice received by the sound receiving module. The request for the inquiry message is transmitted to the servo unit via the communication network.

When the servo unit receives the request for the number from the user end, the servo unit converts the voice query message into an original text content by using a voice recognition technology, and analyzes the original text content by using a semantic analysis technique to obtain an analysis text. Content, and determine whether there is a target problem data portion in all the problem data portions stored in the service database, and the content content has a highest degree of compliance with a predetermined threshold value.

When the SERVOPING unit determines that there is a target problem data part in all the problem data portions stored in the service database, one of the answer data portions stored in the service database corresponds to the answer data part of the target problem data part. As a target answer data part, and at least according to the target answer data part and the current service number of the target service organization, generating a reply message message including at least one service serial number, and determining the stored information in the service database When there is no target problem data part in all the problem data sections, an error message indicating that the voice inquiry message is unrecognizable is generated, and the generated reply message or the error message is generated in response to the request for the number. The first communication network is transmitted to the user end.

When the processing unit of the user terminal receives the reply message or the error message from the server unit by the communication unit, the processing unit controls the user interface unit to regenerate the message of replying the number or the error. a message to enable the user to obtain the retrieved reply message or the error message on at least one of hearing and visual.

The effect of the present invention is that the servo unit can recognize the voice inquiry message obtained by the user end from the user and related to the desired service item by using the voice recognition technology, and use the semantic analysis technology to store the voice inquiry message from the service database. All the question and answer data extracts the target answer data portion with the highest degree of conformity with the voice query message, and provides the target answer data portion and a service serial number to the user end, so that the user can Easily get the correct response message that matches the inquiry message and ensure the correct number for the desired service item.

Referring to FIG. 1, an embodiment of the voice-taking system 100 of the present invention can be applied to, for example, a banking institution having a plurality of branches, wherein the branches can be regarded as a plurality of service organizations, and Providing a customer with a financial transaction or application to any service provider to provide a numbering service prior to the execution of the desired financial transaction or application, wherein the desired financial transaction or application may be vested in the banking institution a service item. The voice taking system 100 includes a plurality of user terminals 1 (only one is depicted in FIG. 1) and a servo unit 2. It is worth noting that, in actual use, the same service organization may allow one or more user terminals to exist, and therefore, each user terminal 1 is located in a corresponding service organization. Hereinafter, only one user terminal 1 (FIG. 1) located in a corresponding service organization (hereinafter referred to as a target service organization) will be described in detail.

In this embodiment, for example, the user terminal 1 is disposed at the target service organization, that is, the user terminal 1 is provided by the target service organization as a branch and is disposed at the target service organization. For use by a user or other user's customers, but not limited to this. In other embodiments, the user 1 can also be a mobile device such as a smart phone or a tablet computer carried by a user who wants to handle the service business and enters the target service organization. The user terminal 1 includes a user interface unit 11, a communication unit 12, and a processing unit 13. The user interface unit 11 includes a radio module 111, a display module 112, and a sound reinforcement module 113. The communication unit 12 is connected to a first communication network 200. The processing unit 13 is electrically connected to the user interface unit 11. And the communication unit 12. In addition, in the case that the user terminal 1 is provided by the target service organization, the first communication network 200 is, for example, an internet network, and the processing unit 13 can also be electrically connected to a column via a connection interface (not shown). Printing device 3, and in the case where the user terminal 1 is a mobile device, the communication unit 12 supports wireless communication and the first communication network 200 is a wireless communication network, and the printing device 3 shown in FIG. 1 can Was omitted.

The servo unit 2 includes a first servo end 21 and a second servo end 22. For example, a banking server owned by the banking institution can be used as the first server 21, and a cloud server commissioned by the banking institution to perform related processing can be used as the second server 22, but not in the network. This limit.

The first server end 21 is provided with a regional database DB1. The regional database DB1 stores a plurality of regional data D1 respectively corresponding to the service organizations, and each regional data D1 includes a plurality of related data. One of the service agencies corresponds to the regional data portion d11 of the service provided by the service agency. The first server end 21 is connected to the user terminal 1 via the first communication network 200 and connected to a second communication network 300, such as the Internet, and, if necessary, to a numbering machine provided at the target service organization. (not shown) maintain communication in order to obtain the current number of the target service organization.

The second server 22 is configured with a service database DB2. The service database DB2 stores a plurality of question and answer data D2 respectively corresponding to a plurality of different service items, and each question and answer data D2 includes a related information. One of the service items corresponds to the problem data portion d21 of the service item, and an answer data portion d22 indicating the corresponding service item. The second server 22 is connected to the first server 21 via the second communication network 300. For example, such service items are related to banking services such as depository, foreign exchange, service desks, large cashiers, general receipts and payments.

Hereinafter, how the voice taking system 100 performs a voice taking procedure will be described in detail with reference to FIGS. 1 through 4. The voice numbering procedure includes the following steps.

First, in step S41, when the radio module 111 of the user terminal 1 receives a voice inquiry message from the user and related to a desired service item, the processing unit 13 uses the communication unit 12 to A request for receiving the voice inquiry message received by the radio module 111 and the organization information of the target service organization is transmitted to the first server 21 via the first communication network 200. It should be noted that, in the case that the user end 1 is provided by the target service organization, the organization information is pre-stored in the first server end 21 (not shown), and at the user end 1 In the case of a mobile device, the user terminal 1 generates the institutional data by executing, for example, an application, and the input operation of the user by the user, but not limited thereto.

Then, in step S42, the first server 21 receives an identification request from the user terminal 1 and transmits an identification request including the voice inquiry message to the second communication network 300. The second servo terminal 22.

Next, in step S43, the second server 22 converts the voice query message into an original text content by using a voice recognition technology when receiving the identification request from the first server terminal 21, and analyzes the semantic content using a semantic analysis technique. The original text content is used to obtain an analytical text content. For example, the original text content is "even deposit" and the analysis text content is "I want to deposit". In another example, the original text content and the analysis text content are "I want to deposit".

Next, in step S44, the second server 22 determines whether there is a target problem data portion d21 in all the problem data portions d21 stored in the service database DB2, and the target problem data portion d21 and the analysis text content have A maximum conformance that is not less than a predetermined threshold. If the result of the determination is affirmative, the flow proceeds to step S45, otherwise the flow proceeds to S46.

In step S45, when the second server 22 determines that there is a target problem data portion d21 in all the problem data portions d21 stored in the service database DB2, all the answer data portions d22 stored in the service database DB2 are obtained. One of the answer data portions d22 corresponding to the target question data portion d21 serves as a target answer data portion d22, and generates a successful identification reply including the analysis text content and the target answer data portion d22. For example, if the analysis text content is "I want to deposit" and has a highest degree of compliance with the text value of a question data portion (target problem data portion) d21, which corresponds to the highest threshold value, The text content of the target answer data portion of the target question data portion d21 is, for example, "deposit class number 1".

In step S46, the second server 22 generates a failure identification reply when it is determined that there is no target problem data portion in all the problem data portions d21 stored in the service database DB2.

Following the steps S45 and S46, the second server 22 transmits the successful identification reply or the failure identification reply to the first server 21 via the second communication network in response to the identification request (step S47).

Then, in step S48, the first server 21 stores the analysis text content for subsequent analysis when receiving the successful identification reply from the second server terminal 22, and according to the target answer data portion d22 One of the pieces of regional data stored in the regional database DB1 corresponds to the regional data portion d11 included in the regional data D1 of the target service organization, and a corresponding portion of the target answer data portion d22 is extracted. The target regional data portion d11 is obtained by, for example, communicating with a destination machine set to the target service organization and related to the target answer data portion d22 (that is, the desired service item). After the service serial number of the current number of the service organization, a reply message including the target regional data portion d11 and the service serial number is generated. For example, according to the above example, if the current number of the bank branch of the target service organization related to the deposit business is "003" and the counter for the deposit business is located on the first floor, the reply message will be sent. Indicates, for example, "Hello, please contact the counter on the first floor. Your number is 004". On the other hand, when receiving the failed identification reply from the second server 22, the first server 21 generates an error message indicating that the voice inquiry message is unrecognizable.

Then, in step S49, the first server 21 transmits the generated reply message or the error message to the user terminal 1 via the first communication network 200 in response to the request for the number.

Finally, in step S50, when the processing unit 13 of the user terminal 1 receives the pickup response message or the error message from the first server 21 by the communication unit 12, the processing unit 13 controls The user interface unit 11 reproduces the take-up reply message or the error message, so that the user can obtain the take-up reply message or the error message on at least one of the hearing and the visual. More specifically, the take-up reply message or the error message may be displayed on the display module 112 and/or reproduced by the sound reinforcement module 113. In addition, when the user terminal 1 is disposed in the target service organization, the processing unit 13 can also control the printing device 3 according to the user's selection, and print out the reply message or the error message. .

In summary, since the voice-taking system 100 of the present invention only needs to receive the voice inquiry message from the user, the voice-receiving and semantic analysis operation of the server unit 2 can be performed, and the service database DB2 and the area are utilized. All the data stored in the database DB1 generates a reply message containing the correct number (service number) and related regional guidelines (target regional data part), and the user terminal 1 is also regenerated and/or The answer reply message 1 is displayed to facilitate the user to obtain the reply message. Therefore, it is possible for the user to achieve the object of the present invention not only because it is extremely convenient to use but also to ensure correct numbering.

However, the above is only the embodiment of the present invention, and the scope of the invention is not limited thereto, and all the simple equivalent changes and modifications according to the scope of the patent application and the patent specification of the present invention are still Within the scope of the invention patent.

100‧‧‧Voice taking system

1‧‧‧User side

11‧‧‧User interface unit

111‧‧‧ Radio Module

112‧‧‧Display module

113‧‧‧Amplifier module

12‧‧‧Communication unit

13‧‧‧Processing unit

2‧‧‧Servo unit

21‧‧‧First server

DB1‧‧‧Regional Database

D1‧‧‧Regional information

D11‧‧‧Regional data section

22‧‧‧Second servo

DB2‧‧‧ Service Database

D2‧‧‧Questions and answers

D21‧‧‧Question data section

D22‧‧‧answer data section

3‧‧‧Printing device

S41-S50‧‧‧Steps

Other features and advantages of the present invention will be apparent from the embodiments of the present invention. FIG. 1 is a block diagram illustrating an embodiment of the voice taking system of the present invention. FIG. 2 is a schematic view. An example of a regional database built by a first server in the embodiment is shown in FIG. 3; FIG. 3 is a schematic diagram showing a service built by a second server in the embodiment. The database; and Figure 4 is a flow chart illustrating, by way of example, how the embodiment performs a voice retrieval procedure.

Claims (12)

A voice numbering method is implemented by a voice numbering system, which includes a user terminal located at a target service organization and a server unit connected to the user terminal via a first communication network. The voice taking method includes the following steps: (A) at least one service database is constructed by the server unit, and the service database stores a plurality of questions and answer data respectively corresponding to a plurality of different service items, each question And the answer material includes a question data portion related to one of the service items of the service item, and an answer data portion indicating the corresponding service item; (B) receiving, by the user end, from a use And related to a voice inquiry message of a service item, transmitting a request for at least the voice inquiry message to the server unit via the first communication network; (C) receiving the message unit by the server unit When the request for the number is received from the user end, the voice query message is converted into an original text content by using a voice recognition technology, and the term is used. The analysis technology analyzes the text content to obtain an analysis text content, and determines whether there is a target problem data portion in all the problem data portions stored in the service database, and the content of the analysis text has a predetermined threshold not lower than a predetermined threshold. (D) by the SERVOPING unit, when it is determined that there is a target problem data part in all the problem data portions stored in the service database, all the answer data stored in the service database is partially included therein. An answer data portion corresponding to the data portion of the target question is used as a target answer data portion, and at least a response message including at least one service serial number is generated according to the target answer data portion and the current access status of the target service organization. And when it is determined that there is no target problem data part in all the problem data portions stored in the service database, an error message indicating that the voice inquiry message is unrecognizable is generated, and the generated message is generated in response to the request for the number. The reply message or the error message, via the first The network is transmitted to the user terminal; and (E) the user terminal regenerates the reply message or the error message when receiving the reply message or the error message from the server unit. So that the user can obtain the reply message or the error message on at least one of the hearing and the visual. The voice taking method according to claim 1, wherein: in the step (A), the server unit further constructs a regional database, where the regional database stores a plurality of regional data respectively corresponding to the plurality of service organizations. Each regional data includes a plurality of regional data portions respectively related to a plurality of service items provided by one of the service agencies; and in step (D), the server unit is further based on the target In the answer data section, one of the regional data included in the regional information corresponding to the target service organization extracts a target regional data portion corresponding to the target answer data portion, so that The take-up reply message generated by the SERVOPACK includes not only the service serial number but also the target regional data portion. The method of claim 2, the server unit includes a first server connected to the user terminal via the first communication network, and a second server connected to the first server via a second communication network. End, wherein: in step (A), the first server constructs the area database, and the second server constructs the service database; in step (B), the number request is transmitted to The first server, and the number request further includes the organization information indicating the target service organization; the step (C) includes the following sub-step (C1), by the first server, receiving the user terminal When the request is received, an identification request including the voice inquiry message is transmitted to the second server via the second communication network, and (C2) is received by the second server from the first When the identification request is made by the server, the voice query message is converted into the original text content by using a voice recognition technology, and the original text content is analyzed by a semantic analysis technique to obtain the analyzed text content, and the determination is performed; Step (D) includes the following sub-step (D1), by the second server, when it is determined that all the problem data portions stored in the service database have the target problem data portion, generating an analysis text content and A successful identification reply of the target answer data portion, and when it is determined that there is no target problem data portion in all the problem data portions stored in the service database, a failure identification reply is generated, and in response to the identification request, The successful identification reply or the failure identification reply is transmitted to the first server via the second communication network, and (D2) is stored by the first server when receiving the successful identification reply from the second server. The analyzing the text content, and extracting the target regional data portion from the regional data portion included in the corresponding regional data according to the target answer data portion, and according to the target regional data portion and the target service The current number of the organization, the response message is generated, and the failure is recognized back from the second server. The error message is generated, and (D3), by the first server, in response to the request for the number, the generated reply message or the error message is transmitted to the use via the first communication network. End. The voice numbering method of claim 3, wherein in step (D2), the first server is further in communication with a numbering machine disposed in the target service organization and corresponding to the desired service item, Obtaining the service serial number generated from the numbering machine and generated according to the current numbering condition. A voice access system includes: a user terminal, located in a target service organization, and including a user interface unit, including a radio module, a communication unit, a first communication network, and a processing unit, Connecting the user interface unit and the communication unit; and a servo unit connected to the first communication network, and configured with a service database, the service database storing a plurality of questions and answers respectively corresponding to a plurality of different service items The information, each question and answer data includes a question data portion related to one of the service items of the service item, and an answer data portion indicating the corresponding service item; wherein, the audio module of the user end Receiving a voice inquiry message from a user and related to a desired service item, the processing unit uses the communication unit to receive a request for at least the voice inquiry message received by the audio module via the a communication network is transmitted to the servo unit; wherein, when the servo unit receives the from the user end When the request is taken, the servo unit converts the voice query message into an original text content by using a voice recognition technology, and analyzes the original text content by using semantic analysis technology to obtain an analysis text content, and determines all the stored contents in the service database. Whether there is a target problem data part in the problem data part, and having the highest degree of compliance with the analysis text content not lower than a predetermined threshold value; wherein the server unit determines all the problems stored in the service database When there is a target problem data part in the data part, one part of all the answer data parts stored in the service database corresponds to the answer data part of the target question data part as a target answer data part, and at least according to the target answer data part And the current numbering situation of the service organization, generating a reply message containing at least one service serial number, and generating a message when it is determined that there is no target problem data part in all the problem data portions stored in the service database Indicates that the voice inquiry message cannot be identified And the response message is sent to the user end via the first communication network, and the processing unit of the user end is sent to the user terminal; When the communication unit receives the retrieved reply message or the error message from the server unit, the processing unit controls the user interface unit to regenerate the retrieved reply message or the error message, so that the user can enable the user to Obtaining the reply message or the error message on at least one of hearing and vision. The voice taking system of claim 5, wherein: the server unit further comprises a regional database, wherein the regional database stores a plurality of regional data respectively corresponding to the plurality of service organizations, and each regional data includes a plurality of regional data portions respectively related to a plurality of service items provided by one of the service agencies; and the server unit further corresponding to one of the regional data according to the target answer data portion Extracting a target regional data portion corresponding to the target answer data portion of the regional data portion of the target service organization, so that the response message generated by the server unit includes not only the The service serial number also contains the target regional data section. The voice taking system of claim 6, wherein: the request for the number from the user terminal further includes the organization information indicating the target service organization; the server unit includes a first server, and the area information is configured. The library is connected to the user terminal via the first communication network to receive the request for the number from the user terminal, and is connected to a second communication network, and a second server is configured with the service database, and Connecting the first server to the second server; the first server receives an identification request from the user terminal, and transmits an identification request including the voice inquiry message to the second communication network. Up to the second server; the second server receives the identification request from the first server, converts the voice query message into the original text content by using a voice recognition technology, and analyzes the original by semantic analysis technology Text content to obtain the analysis text content, and perform the determination, and determine that all the problem data stored in the service database is stored In the data part of the target question, a successful identification reply containing the content of the analysis text and the data part of the target answer is generated, and in the data part of all the problem data stored in the service database, there is no target problem data part. And generating a failure identification reply, and transmitting the successful identification reply or the failure identification reply to the first server via the second communication network in response to the identification request; and the first server receives the And storing the analysis text content on the successful recognition reply of the second server, and extracting the target regional data portion from the regional data portion included in the corresponding pen regional data according to the target answer data portion, and according to the The target regional data portion and the current number of the target service organization generate the response message, and when the failure identification response from the second server is received, the error message is generated, and the response is received. The request message and the generated reply message or the error message are sent via the first communication network The road is transmitted to the user end. The voice taking system of claim 7, wherein the first server is further in communication with a numbering machine disposed in the target service organization and corresponding to the desired service item, to obtain the The service serial number generated according to the current number taking condition. The voice access system of claim 5, wherein the user interface unit of the user terminal further comprises a display module for displaying the number response message or the error message, and a voice mode Regenerating the sound reinforcement module that takes the reply message or the error message. The voice-taking system of claim 5, wherein the user terminal is provided by the target service organization and is disposed in the target service organization, and the first communication network is identical to the second communication network. The voice taking system of claim 10, further comprising a printing device disposed in the target service mechanism and electrically connected to the processing unit of the user end, the processing unit controlling the printing device to selectively print The reply message is sent. The voice access system of claim 5, wherein the user terminal is a mobile device, and the first communication network is a wireless communication network.