TWM579322U - Voice serving system - Google Patents

Abstract

A voice service system includes a call device, a semantic recognition system, a customer identity database, a voiceprint recognition system, an interactive voice response system, and a computer host. The interactive voice response system first requires the calling device to input a voice data by a user, and the computer host transmits the voice data to the semantic recognition system to obtain the personal identity information. When the personal identity information is compared and matched with the personal identity information stored in the customer identity database, the computer host transmits the voice data to the voiceprint recognition system for voiceprint recognition, and corresponds to the voice The voice input data of the data is compared with the voiceprint data stored in the database, and the identity of the user is determined.

Description

Voice service system

The present invention relates to a voice service system, and more particularly to a voice service system for performing semantic analysis and voiceprint analysis on a single voice data to verify identity.

Existing financial players, such as banks or insurance companies, generally provide an interactive voice response (IVR) service for voice-based value-added services. After the client establishes a connection with an interactive voice response system providing an interactive voice response service by a call device, the interactive voice response system often asks the client to input a person profile by the call device, for example, by pressing a button The identity card number is entered to obtain the identity of the customer corresponding to the identity card number, and then the predetermined related voice service is provided according to the identity of the client. However, the conventional interactive voice response system does not completely recognize the identity of the client. For example, other users use the identity card number of the customer that has been obtained to fraudulently use the client. Therefore, how to provide more Correct identification of the customer's identity becomes a problem to be solved.

Therefore, the purpose of the present invention is to provide a voice service system that performs semantic analysis and voiceprint analysis on a single voice material to verify identity.

Thus, the present invention provides a voice service system comprising a call device, a semantic recognition system, a customer identity database, a voiceprint recognition system, an interactive voice response system, and a computer host.

The call device is adapted for input by a user to generate a voice material, the voice message including personal identity information related to the user. The semantic recognition system performs semantic recognition. The customer identity database pre-stores the personal identity information corresponding to the user. The voiceprint recognition system includes a voiceprint comparison database that stores a voiceprint data corresponding to the user, and performs voiceprint recognition based on the voiceprint data. An interactive voice response service is provided when the interactive voice response system is connected to the call device.

The computer host electrically connects the interactive voice response system, the semantic recognition system, the customer identity database, and the voiceprint recognition system. When the interactive voice response system provides an interactive voice response service to the calling device, the calling device is first requested to input the voice data by the user, and the voice data is transmitted to the computer host. The computer host transmits the voice data to the semantic recognition system to obtain the personal identity information, and compares the personal identity information with the personal identity information stored in the customer identity database.

When the personal identity information is matched, the computer host transmits the voice data to the voiceprint recognition system for voiceprint recognition, and the voiceprint input data corresponding to the voice data is compared with the voiceprint ratio. After the voiceprint data stored in the database is met, the computer host determines that the identity of the user is the personal identity information.

In some implementations, wherein the computer host determines that the identity of the user is corresponding to the personal identity information, the interactive voice response system is notified to provide a first type of interactive voice response service to the calling device.

In some implementations, wherein the first type of interactive voice response service comprises a financial query service.

In other implementations, the voice service system further includes a call terminal electrically connected to the host computer. Wherein, when the result of the personal identity information comparison is not met, or when the result of the voiceprint comparison is not met, the computer host notifies the call terminal, and a customer service specialist uses the call terminal and The calling device manually confirms the personal identity information of the user with the user.

In some implementations, after the computer host determines that the user's identity is corresponding to the personal identity information, or after the customer service specialist and the user manually confirm the personal identity information of the user, The computer host transmits a One Time Password (OTP) to the calling device as a short message, and notifies the interactive when the one-time password input by the calling device is received and the judgment is met. The voice response system provides a second type of interactive voice response service to the call device.

In some implementations, wherein the second type of interactive voice response service comprises a financial transaction service.

In other implementations, the personal identity information includes at least one of a personal identification number, a name, a date of birth, and a mobile number corresponding to the user.

The function of the present invention is to perform a semantic analysis on the voice data generated by the communication device by the semantic recognition system, and first confirm that the personal identity information of the user is related to the personal identity information stored in the customer identity database. Compatible, and then the voiceprint recognition system performs voiceprint recognition on the same voice data, and then confirms that the voiceprint data stored in the voiceprint comparison database matches the voiceprint data to achieve double confirmation of the user's Personal identity information, and can more accurately identify the identity of the customer.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same reference numerals.

Referring to FIG. 1, an embodiment of the present voice service system includes a call device 1, a semantic recognition system 4, a customer identity database 5, a voiceprint recognition system 6, an interactive voice response system 2, and a computer. The host 3 and a call terminal 7. The voice service system is provided by a financial provider, and in this embodiment, is provided by an insurance company, but is not limited thereto.

The semantic recognition system 4 can perform semantic recognition on the received voice material by using the prior art, and obtain the semantic content included in the voice data. The customer identity database 5 pre-stores a plurality of personal identity information corresponding to a plurality of customers of the insurance company. In this embodiment, the personal identity information is a personal identification number corresponding to the customer, and in other implementations, the The personal identity information may also be at least one of the identity card number, a name, a date of birth, and a mobile number corresponding to the customer.

The voiceprint recognition system 6 includes a voiceprint comparison database 61 for storing a plurality of voiceprint data of the clients, and can perform voiceprint recognition based on one of the voiceprint materials by the prior art. The interactive voice response (IVR) system can provide an interactive voice response service to the call device 1 when the call device 1 establishes a connection.

The communication device 1 is adapted to be input by a user to generate the voice data, the voice data including the personal identity information related to the user. In this embodiment, the communication device 1 is a smart phone, and the user makes a call by the smart phone, so that the communication device 1 establishes a connection with the interactive voice response system 2. In other embodiments, the communication device 1 can also be a conventional indoor telephone, or other device that can implement a voice service provided by a carrier, and the like.

The computer host 3 is electrically connected to the interactive voice response system 2, the semantic recognition system, the customer identity database 5, and the voiceprint recognition system 6. After the communication device 1 is connected to the interactive voice response system 2, the interactive voice response system 2 provides an interactive voice response service to the communication device 1 and first requests the communication device 1 to input by the user. The voice material. For example, the interactive voice response system 2 transmits a voice message "Good morning / good afternoon / good night, Taiwanese life is good, please say that the identity card number of the insurance person is ten yards" is transmitted to the communication device 1 to The user is provided to listen to the voice message, and then input the voice data.

Then, the interactive voice response system 2 transmits the voice data to the computer host 3, and the computer host 3 transmits the voice data to the semantic recognition system 4, and semantically identifies the voice data to obtain the personal identity. The information is further compared with the personally identifiable information of the customers stored in the customer identity database 5.

When the result of the personal identity information comparison is consistent with one of the customers, the computer host 3 transmits the voice data to the voiceprint recognition system 6 for voiceprint recognition, and corresponding to the voice data. After the voice input data matches the voiceprint data of the one of the customers stored in the database 61, the computer host 3 determines that the user's identity is corresponding to the customer. The personal identity information of the person.

When the host computer 3 determines that the user's identity is corresponding to the personal identity information, the interactive voice response system 2 is notified to provide the first type of interactive voice response service to the communication device 1. In this embodiment, the first type of interactive voice response service includes a financial inquiry service, for example, querying policy related information that the user has insured at the insurance company, insurance product information currently being sold, and the like.

In another aspect, when the personal identity information comparison result is not met, or when the result of the voiceprint data comparison is not met, the computer host 3 notifies the call terminal 7 and a customer service The caller 7 and the call device 1 manually confirm the personal identity information of the user with the user, that is, the personal information of the user is verbally asked to be stored with the customer identity database 5 The data is used for manual inquiry and comparison.

After the computer host 3 determines that the user's identity is corresponding to the personal identity information, or after the customer service specialist and the user manually confirm the personal identity information of the user, the computer host 3 will once A One Time Password (OTP) is transmitted to the communication device 1 as a short message. After the user obtains the one-time password by the communication device 1, the user can input and generate the one-time password through the communication device 1, for example, by means of button input, voice input, or message transmission, so that the computer host 3 is obtained. The one-time password from the communication device 1 and notifying the interactive voice response system 2 when determining that the one-time password from the communication device 1 matches the one-time password generated by the computer host 3 The call device 1 provides a second type of interactive voice response service. In this embodiment, the second type of interactive voice response service includes a financial transaction service, such as online purchase of insurance products, online change of insurance contract content, and the like.

Referring to FIG. 1 and FIG. 2, FIG. 2 is a voice service method performed by the voice service system of the present invention. As can be seen from the foregoing embodiment of the present novel voice service system, the voice service method includes the following steps S1 to S8.

In step S1, the voice data is generated by the user input, and the voice data includes the personal identity information related to the user.

In step S2, semantic analysis is performed on the voice data to obtain the personal identity information, and the personal identity information is compared to determine whether the content is consistent.

In step S3, when the result of the personal identity information comparison is matched, the voice data is identified and the voiceprint is compared to determine whether the match is met.

In step S4, when the result of the voiceprint comparison is consistent, that is, the comparison made by the personal identity information is also consistent, it is determined that the identity of the user is the corresponding personal identity information.

In step S5, the first type of interactive voice response service is provided.

In step S6, when the result of the personal identity information comparison is not met, or when the result of the voiceprint comparison does not match, the customer service specialist and the user manually confirm the user's Personal identity information.

In step S7, after the computer host 3 determines that the user's identity is corresponding to the personal identity information, or after the customer service specialist and the user manually confirm the personal identity information of the user, the primary occurrence is generated. The password is used to notify the user and receives the one-time password entered by the user. In addition, in this embodiment, the host computer 3 generates and transmits the one-time password by using a one-time password (OTP) transmission system, and in other embodiments, the one-time password is also After the customer service specialist and the user manually confirm the personal identity information of the user, the customer service specialist operates the one-time password (OTP) transmission system to generate and transmit the one-time password, and the one-time password The password will not be known to the customer service specialist.

In step S8, the second type of interactive voice response service is provided when it is determined that the generated one-time password is met. In addition, in this embodiment, it is determined by the host computer 3 whether the one-time password is consistent, and in other embodiments, when the user replies to the customer service specialist by telephone dictation The customer service specialist then enters the one-time password into the one-time password transmission system to verify compliance.

In summary, the present invention uses the semantic recognition system to perform semantic analysis on the voice data generated by the communication device, and then the voiceprint recognition system performs voiceprint recognition on the same voice data to double confirm the pair. The user's personal identity information should be used to more accurately identify the customer's identity. In addition, the computer host can also provide two types of interactive voice response services, such as financial inquiry service and financial transaction service, by using the one-time password verification method, so that the purpose of the novel can be achieved.

However, the above is only the embodiment of the present invention. When the scope of the novel implementation cannot be limited thereto, all simple equivalent changes and modifications according to the scope of the patent application and the contents of the patent specification are still This new patent covers the scope.

1‧‧‧Call device  2‧‧‧Interactive voice response system  3‧‧‧Computer host  4‧‧‧Speech Identification System  5‧‧‧Customer Identity Database  6‧‧‧ Voiceprint Identification System  61‧‧‧ voiceprint comparison database  7‧‧‧Call terminal  S1~S8‧‧‧Steps

Other features and effects of the novel will be apparent from the embodiments of the drawings, in which:  1 is a block diagram showing an embodiment of the present novel voice service system; and  2 is a flow chart illustrating a voice service method performed by the embodiment.  

Claims (7)

A voice service system comprising:  a call device adapted to be input by a user to generate a voice data, the voice data including personal identity information related to the user;  A semantic recognition system that performs semantic recognition;  a customer identity database for pre-storing the personal identity information corresponding to the user;  a voice pattern recognition system, comprising: a voiceprint comparison database storing a voiceprint data corresponding to the user, and performing voiceprint recognition according to the voiceprint data;  An interactive voice response system that provides an interactive voice response service when establishing a connection with the call device;  a computer host electrically connecting the interactive voice response system, the semantic recognition system, the customer identity database, and the voiceprint recognition system,  Wherein, when the interactive voice response system provides an interactive voice response service to the calling device, the calling device is first requested to input the voice data by the user, and the voice data is transmitted to the computer host, the computer host Transmitting the voice data to the semantic recognition system to obtain the personal identity information, and comparing the personal identity information with the personal identity information stored in the customer identity database.  When the personal identity information is matched, the computer host transmits the voice data to the voiceprint recognition system for voiceprint recognition, and the voiceprint input data corresponding to the voice data is compared with the voiceprint ratio. After the voiceprint data stored in the database is met, the computer host determines that the identity of the user is the personal identity information.   The voice service system of claim 1, wherein, after the computer host determines that the identity of the user is the personal identity information, the interactive voice response system is notified to provide the first type of interactive voice to the communication device. Answer service. The voice service system of claim 2, wherein the first type of interactive voice response service comprises a financial inquiry service. The voice service system of claim 2, further comprising a call terminal electrically connected to the host computer, wherein when the result of the personal identity information comparison is not met, or when the voiceprint data is compared, the result is When it is not met, the computer host notifies the call terminal, and a customer service specialist manually confirms the personal identity information of the user with the user through the call terminal and the call device. The voice service system of claim 4, wherein the computer host determines that the identity of the user is corresponding to the personal identity information, or when the customer service specialist and the user manually confirm the user's After the personal identity information, the computer host transmits a One Time Password (OTP) to the communication device by way of a short message, and when receiving the one-time password input by the communication device and determining that the correspondence is met, The interactive voice response system is notified to provide a second type of interactive voice response service to the call device. The voice service system of claim 5, wherein the second type of interactive voice response service comprises a financial transaction service. The voice service system of claim 1, wherein the personal identity information includes at least one of a personal identification number, a name, a date of birth, and a mobile number corresponding to the user.