KR20040016182A - Method and system for verifying user - Google Patents

Abstract

PURPOSE: A method and an apparatus for authenticating a user are provided to authenticate a user by using voice features of the user. CONSTITUTION: A terminal receives an audio signal spoken by a user(601). An authentication server receives the audio signal and extracts a speaker component and feature elements(603,605). The authentication server executes an authentication process by using the extracted feature elements(607). The authentication server judges whether the extracted feature elements are valid(609). The authentication server generates a permission message or a rejection message according to the result of the judgement(611,613). The generated message is transmitted to the terminal.

Description

User authentication method and device {METHOD AND SYSTEM FOR VERIFYING USER}

The present invention relates to a user authentication method and apparatus.

With the development of computers, society has become increasingly complex and requires the rapid exchange of information. In particular, in recent years, the use of credit cards has explosively increased, and accordingly, there has been an increase in the number of illegal and unjust purchases of goods or services using credit cards. In order to prevent unauthorized persons from using the credit card of another person to save goods or services, the merchant checks or copies the cardholder's identification card or photographs it using a recording medium such as CCTV and stores it. However, the above acts are performed only in large affiliated stores such as department stores. Small merchants are conducting credit card transactions with the approval of bans (VANs) without undergoing identity verification due to poor circumstances or consumer complaints. Therefore, a small merchant who does not store a recording medium has a problem in that it is necessary to reimburse the damage caused by the use of the card if it is found that the credit card approved by the transaction is used by an unauthorized person due to theft or loss.

Not only transactions using credit cards, but also transactions using debit cards and electronic money require a series of processes to verify and authenticate the transaction by a genuine user.

Accordingly, an object of the present invention is to solve the above problems.

A specific object of the present invention is to provide a user authentication method and apparatus capable of authenticating a user by using a voice characteristic among the unique characteristics of each individual without having to install additional equipment for user verification.

Another object of the present invention is to provide a user authentication method and apparatus that can easily check whether the credit card holder is a legitimate user without the need for expensive equipment such as CCTV in a small credit card merchant.

It is still another object of the present invention to provide a user authentication method and apparatus capable of confirming or authenticating whether a user is a true user in a transaction using a debit card and electronic money.

It is still another object of the present invention to provide a user authentication method and apparatus for improving a speech recognition rate by adding a natural language processing function, a numeric sound processing function, a keyword detection function, and a noise processing function.

1 is a schematic configuration diagram of a system to which the present invention can be applied.

2 is a flowchart schematically illustrating a process of registering voice information of a user according to an exemplary embodiment of the present invention.

3 is a schematic flowchart showing a preprocessing process according to an exemplary embodiment of the present invention.

4 illustrates a distributed processing scheme in accordance with one preferred embodiment of the present invention.

5 is a flowchart schematically showing a registration process according to an embodiment of the present invention.

6A is a flowchart illustrating a user authentication process according to an exemplary embodiment of the present invention.

6B is a flowchart illustrating a user authentication process according to another exemplary embodiment of the present invention.

7 is a view schematically showing a recognition process according to an embodiment of the present invention.

8 is a view schematically showing the configuration of a voice signal preprocessing system according to an embodiment of the present invention.

9 is a flowchart schematically illustrating a process of approving a credit card transaction according to an exemplary embodiment of the present invention.

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

101... Terminal 103. Authentication server

105... PG company 107. VAN company

109... Credit card company

In order to achieve the above object, according to an aspect of the present invention, in a user authentication method in a credit card transaction, receiving a voice signal uttered by the user from a merchant terminal, the pre-processing process from the received voice signal Extracting the speaker component by performing, extracting a feature element from the extracted speaker component, comparing the extracted feature element with a pre-set feature element stored in a storage device, and determining whether the extracted feature element is valid. And generating a message corresponding to the determination result and transmitting the generated message to the affiliated store terminal, wherein the affiliated store terminal performs a credit card transaction using the message. How to authenticate users in credit card transactions There.

In a preferred embodiment, the merchant terminal preferably receives the voice signal through a microphone provided therein. The method may further include converting the voice signal into a digital signal.

According to another aspect of the present invention, in a user authentication method in a credit card transaction, receiving a speaker component extracted from a voice signal uttered by the user from a merchant terminal, extracting a feature element from the received speaker component Comparing the extracted feature elements with pre-configured feature elements stored in a storage device, determining whether the extracted feature elements are valid, generating a message corresponding to the determination result, and generating the generated message. Including the step of transmitting to the merchant terminal, the merchant terminal may provide a user authentication method in a credit card transaction, characterized in that for performing a credit card transaction using the message.

In a preferred embodiment, the merchant terminal may extract the speaker component from the voice signal by performing a preprocessing process, and preferably receive the voice signal through a microphone provided therein. In addition, the terminal preferably converts the voice signal into a digital signal.

According to another aspect of the present invention, in the authentication device for authenticating a user in a credit card transaction, means for receiving a voice signal uttered by the user from a merchant terminal, by performing a preprocessing process from the received voice signal Means for extracting speaker components, means for extracting feature elements from the extracted speaker components, means for comparing the extracted feature elements with pre-set feature elements stored in a storage device, means for determining whether the extracted feature elements are valid; Means for generating a message corresponding to the determination result and means for transmitting the generated message to the merchant terminal, wherein the merchant terminal performs a credit card transaction using the message. Can provide an authentication device that authenticates users in All.

According to another aspect of the present invention, an authentication apparatus for authenticating a user in a credit card transaction, comprising: means for receiving a speaker component extracted from a voice signal uttered by the user from a merchant terminal, and a feature element from the received speaker component Means for extracting the means, means for comparing the extracted feature elements with pre-set feature elements stored in a storage device, means for determining whether the extracted feature elements are valid, means for generating a message corresponding to the determination result and generating Means for transmitting a message to the merchant terminal, the merchant terminal may provide an authentication device for authenticating a user in a credit card transaction, characterized in that for performing a credit card transaction using the message.

According to another aspect of the present invention, an authentication server for authenticating a user in a credit card transaction, comprising: a memory in which a program is stored, and a processor coupled to the memory to execute the program, wherein the processor is configured to execute the program. Receiving a voice signal uttered by the user from a merchant terminal, extracting the speaker component by performing a preprocessing process from the received voice signal, extracting a feature element from the extracted speaker component, the extracted feature Compare the element with a pre-set feature element stored in the storage device, determine whether the extracted feature element is valid, generate a message corresponding to the determination result, and transmit the generated message to the merchant terminal, the merchant The terminal uses the message to make a credit card transaction Your credit card transactions, which comprises performing an authentication server that can provide authentication.

According to another aspect of the present invention, an authentication server for authenticating a user in a credit card transaction, comprising: a memory in which a program is stored, and a processor coupled to the memory to execute the program, wherein the processor is configured to execute the program. By receiving a speaker component extracted from the voice signal uttered by the user from a merchant terminal, extracting a feature element from the received speaker component, and comparing the extracted feature element with a pre-set feature element stored in a storage device; Determining whether the extracted feature element is valid, generating a message corresponding to the determination result, and transmitting the generated message to the merchant terminal, wherein the merchant terminal performs a credit card transaction using the message. Use in credit card transactions, characterized in that It can provide an authentication server that authenticates the child.

Next, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Figure 1 shows a schematic diagram of a system to which the present invention can be applied.

1, a user authentication system according to an embodiment of the present invention is at least one merchant terminal (hereinafter, referred to as a terminal; 101), authentication server 103, payment agency (hereinafter referred to as PG company; 105) ), A bansa (VAN) 107 and a card company 109. The present invention relates to a method and system that can identify or authenticate a user in a transaction using a credit card, a transaction using a debit card, a transaction using electronic money, and the like. For convenience of explanation, only a transaction using a credit card will be described, and a user authentication method and apparatus using a credit card can be used for a transaction using a debit card and / or a transaction using electronic money without requiring an additional device. It will be apparent to those skilled in the art that the present invention can be applied. In addition, the terminal 101 may be any one of performing a user authentication method according to the present invention, such as a credit card terminal, PDA, mobile communication terminal, home electronic money terminal. However, for convenience of description, the terminal 101 is limited to being a credit card terminal. Based on the user authentication system in a credit card transaction according to the present invention, a credit card transaction process through user authentication will be described as follows.

After a user having a credit card purchases goods and services and wants to pay using the credit card, the terminal 101 receives a voice signal of the user through a microphone provided therein. Thereafter, the terminal 101 extracts the speaker component as it is or through a preprocessing process, and transmits the speaker component to the authentication server 103. The authentication server 103 extracts the speaker component by performing the preprocessing process on the received voice signal or extracts a feature element which is an identification element of the user using the speaker component directly received from the terminal 101. Thereafter, the authentication server 103 performs the user's verification process using the extracted feature elements. The authentication server 103 transmits the performance result back to the terminal 101. If the execution result is determined to be valid, the terminal 101 requests the card payment approval from the PG company 105. The PG company 105 requests the card company 109 to approve the card payment in real time via the bansa 107. The card company 109 uses the credit card to approve transactions at the affiliated store in response to a card payment approval request. Finally, the merchant receives the handwritten signature from the user and keeps the slip.

The authentication server according to an embodiment of the present invention stores the voice information according to the speaker in a DB in advance, and can verify the identity of the credit card user using the voice information DB. Voice information registration of the user in the authentication server according to the present invention can be largely by three methods. First, the authentication server may receive voice information from a user in advance through a communication network, in particular, the Internet, and store the information in a DB through a predetermined registration process. Second, the authentication server can directly record the voice of the user by directing the user to visit directly, and extract voice information therefrom and store it in the DB. Third, the authentication server may copy and use the voice information DB prepared in advance from a financial company such as a card company or a bank. Hereinafter, the authentication server processes the voice information received through the Internet through a certain registration process, and will be described below with reference to the first case of storing it in a DB.

2 is a flowchart schematically illustrating a process of registering voice information of a user according to an exemplary embodiment of the present invention.

Referring to FIG. 2, when a user transmits a connection request signal from a client through an input device, the authentication server receives the connection request signal and then transmits a web document to the client. Here, the client is preferably a general personal computer. Here, the web document is a user inputs a predetermined input, for example, a member identifier, name, gender, address, social security number, e-mail address, credit card number, etc. in the web document for performing a general network service Embedding a web document in the form of a form that can be done. The client receives the web document and displays it on the display device. Thereafter, the user receives the predetermined input items, for example, a member identifier, a name, a gender, an address, a social security number, an e-mail address, a credit card number, and the like through the input device and transmits the same to the authentication server. . Thereafter, the authentication server receives the predetermined input and stores it in an internal database. Thereafter, the authentication server determines a word or the like to be pronounced by the user using the predetermined input item. The authentication server transmits a web document including a word to be pronounced by the user to the client. In addition, the web document includes a component software for preprocessing a voice signal to be pronounced by a user corresponding to the determined word, extracting speaker components corresponding to each word, and transmitting the same to a server. )

The client receives the web document and displays it on a display device, and according to a preferred embodiment of the present invention, receives a voice signal corresponding to the word from a user through a microphone (step 201). Thereafter, the client performs A / D conversion (not shown) on the voice signal, and then performs a preprocessing process using the same as an input. After performing the preprocessing, the client extracts the speaker component (step 203) and transmits it to the authentication server.

3 is a schematic flowchart illustrating a preprocessing process according to an exemplary embodiment of the present invention.

Referring to FIG. 3, a preprocessing process according to an exemplary embodiment of the present invention includes sampling a voice signal of a user received through a microphone (step 301), and removing a silent section from the sampled sound signal (step 303). ), High-band filtering the speech signal from which the silence interval has been removed (step 305), frame blocking the filtered speech signal (step 307), windowing step (309), and extracting the correlation function (step 311). LPC analysis step 313. Through the LPC analysis step, the speaker component (LPC-Cepstrum) is extracted (step 315).

4 is a diagram illustrating a distributed processing method according to an embodiment of the present invention.

Referring to FIG. 4, the speaker authentication system according to the preferred embodiment of the present invention may be largely composed of a client component 403 and a server component 401. The client component 403 consists of two interfaces. It consists of an interface that extracts the speaker component from the digitized voice signal and an interface that is informed of the recognition result from the server. The proxy object performs a marshalling process to packetize the extracted speaker component and then transmits it to the server through a channel and receives voice pattern data from the packet transmitted to the server through a stub object. Perform unmarshalling process to restore and deliver speaker component to recognition process interface in server's recognition component. After performing the recognition process, the result is notified to the result operation interface of the client in the reverse order of the previous procedure.

By such a distributed processing scheme, even when many users access the authentication server at the same time on the Internet, it is possible to prevent the entire system from concentrating the load on some systems.

Referring back to FIG. 2, the authentication server receives a speaker component from the client and then performs a registration process (step 205). The registration process will be described in detail with reference to FIG. 5.

5 is a flowchart schematically illustrating a registration process according to an exemplary embodiment of the present invention.

Referring to FIG. 5, the registration process includes a K-means crusting step 501, a vector quantization step 503, a probability value initialization step 505, a forward backward algorithm step 507, and a Baum-welch estimation step 509. It includes.

Referring back to FIG. 2, the authentication server extracts feature elements from the speaker component by the registration process (step 207). The authentication server then stores the extracted feature elements in a pre- equipped database (step 209).

6A is a flowchart illustrating a user authentication process according to an exemplary embodiment of the present invention.

If the user wants to pay for the purchase of goods or services using a credit card possessed by the user, the terminal receives a voice signal uttered by the user corresponding to a predetermined word through a microphone (step 601). Thereafter, the terminal transmits the received voice signal directly to the authentication server. Here, the terminal is a voice signal after the speech signal pre-processing process, which is a series of processing by the noise processing function, natural language processing function, continuous word recognition function, key word recognition function, numeric sound recognition function, etc. Can be sent to the authentication server. In another embodiment of the present invention, the voice signal preprocessing process may be performed by the authentication server. That is, when the terminal transmits the voice signal uttered by the user to the authentication server, the authentication server performs the sound signal preprocessing process, and then performs a process of extracting a speaker component. The audio signal preprocessing will be described in detail with reference to FIG. 7.

Here, the terminal may further include a voice synthesis module (TTS module) and an automatic answering system (ARS). The speech synthesis module converts text information into voice in real time, and specifically, it is a device that provides sound close to human development by including natural rhyme information by clearly grasping and analyzing grammatical structure. An automatic answering system refers to a device that outputs a procedure to be performed by a user by voice, and outputs a specific result accordingly when the user performs a certain action accordingly (for example, pressing a specific button). Therefore, a user with a credit card can perform a series of actions according to the voices outputted by the merchant terminal using the speech synthesis module and the answering machine, and in particular, a word specific to the microphone (for example, a person's name or Card expiration date, etc.) can be spoken.

After receiving the voice signal corresponding to the user from the terminal, the authentication server extracts the speaker component and the feature element using the voice signal (step 603 and step 605). Extraction of the speaker component and the feature element is the same as described above, and thus a detailed description thereof will be omitted. Thereafter, the authentication server performs a recognition process using the extracted feature element (step 607). The confirmation process will be described in detail with reference to FIG. 7.

7 is a view schematically showing a recognition process according to an embodiment of the present invention.

Referring to FIG. 7, the recognition process includes the speaker component and the feature element as inputs, and are extracted through the VQ indexing step (step 701), the observation parameter extraction step (step 703), and the Viterbi calculation step (step 705). Comparing the feature element with the corresponding stored feature element corresponding to the user, and then determining the result (step 707).

Referring again to FIG. 6A, the authentication server determines whether the extracted feature element is valid (i.e., whether it is pronounced by a legitimate user) by calculating the result of the recognition process (step 609). Check whether the user is a true user. The authentication server generates a grant message or a reject message according to the determination result (steps 611 and 613). Thereafter, the authentication server transmits the generated message to the terminal.

6B is a flowchart illustrating a user authentication process according to another exemplary embodiment of the present invention. Step 615 in FIG. 6B is the same as step 601 in FIG. 6A, and steps 619 through 627 are the same as steps 605 through 613 in FIG. 6A, and thus a detailed description thereof will be omitted.

Referring to FIG. 6B, the terminal preprocesses a voice signal to be pronounced by the user, and stores component software in an internal memory for extracting speaker components corresponding to each word and transmitting them to an authentication server. Doing. The terminal extracts the speaker component through the preprocessing process using the component software on the voice signal received through the microphone (step 617). Since the pretreatment process is as described above, a detailed description thereof will be omitted. Thereafter, the terminal transmits the extracted speaker component to the authentication server.

8 is a diagram schematically showing the configuration of a voice signal preprocessing system according to an embodiment of the present invention.

Referring to FIG. 8, a voice signal preprocessing system may be included in the terminal or the authentication server and includes at least application modules such as a transmission / reception processing module 862, a voice processing module 864, a noise processing module 866, and the like. .

The voice signal preprocessing system can use various operating systems (OS) as the OS of the system. The OS provides a high level command to an application program interface (API) 802 to control the operation of each application module 860.

The voice signal preprocessing system identifies a corresponding application module 860 according to a high level command provided from the API 802, and provides a high level command processing unit 804 for decoding the high level command and providing the same to a corresponding place. Include. The application module controller 850 controls the operation of the application module 860 according to the instructions provided from the high level command processor 804. That is, the high level instruction processor 804 identifies whether there is an application module 860 corresponding thereto according to the high level instruction provided through the API 802, and if the corresponding application module 860 exists, The decoded command is recognized by the application module 860 to be transmitted to a corresponding transmission device or a message transmission is controlled. Accordingly, the application module control unit 850 includes a mapping unit 851, 853, 855 and an interface unit 852, 854, 856 for the transmission / reception processing module 862, the voice processing module 864, and the noise processing module 866. Each).

The transmit / receive processing module mapping unit 851 receives a high level command for performing communication between the authentication server or the terminal from the high-level command processing unit 804, and sends it to a device level that can be processed by the transmit / receive processing module 862. Mapping is provided to the transmission and reception processing module 862 through the transmission and reception processing module interface unit 852.

The speech processing module mapping unit 853 and the speech processing module interface unit 854 perform functions such as natural language processing, continuous word recognition processing, key word detection processing, numeric sound recognition processing, etc., by a user through a microphone. Part. That is, the user may recognize a continuous word, detect a key word, or additionally recognize a numeric sound in the voice signal spoken by the user. In the preferred embodiment of the present invention, the speech signal preprocessing process may be performed only by the natural language processing function, but there is a problem that the recognition rate is about 85% only by the natural language processing. In another preferred embodiment of the present invention, the recognition rate can be improved up to 100% by adding a number recognition function. Specifically, when the user presents a suggestion including a number (for example, two digits of a credit card number), and the user speaks the suggestion, the speech signal preprocessing system recognizes the speech signal more efficiently. can do. The voice processing module mapping unit 853 receives a high level command for using the voice processing module 864 from the high level command processing unit 804 and maps it to a device level command, and the voice processing module interface unit 854 Through the voice processing module 864 is provided.

The noise processing module 866 is a part of processing noise included in a voice signal spoken by a user through a microphone. The noise processing module mapping unit 855 receives the high level command applied through the high level command processing unit 804 and maps it to a device level command that can be recognized by the noise processing module 866. Device level instructions are provided to the noise processing module 866 via the noise processing module interface 856.

9 is a flowchart schematically illustrating an approval process of a credit card transaction according to an exemplary embodiment of the present invention.

Referring to FIG. 9, when it is recognized that the user is a legitimate user based on the message received from the authentication server, the terminal generates a request signal for approval of a credit card transaction and transmits it to the PG company. In response, PG will send an authorization request signal to VAN and credit card companies in real time. In response, the credit card company, which has received the authorization request signal, determines whether the authorization is possible using the credit information of the user, and transmits the authorization signal to the VAN company when there is no defect. The approval signal may be transmitted to the terminal via the PG company. The terminal performs the normal credit card transaction based on the received authorization signal. That is, the merchant performs the transaction by reading the credit card information recorded in the credit card from the user, and finally receives the user's handwritten signature on the card slip.

The present invention is not limited to the above embodiments, and many variations are possible by those skilled in the art within the spirit of the present invention.

As described above, according to the present invention, a user authentication method and apparatus capable of authenticating a user by using voice characteristics among unique characteristics owned by each individual without having to install additional equipment for user verification. Can be provided.

In addition, according to the present invention, it is possible to provide a user authentication method and apparatus that can easily check whether the credit card holder is a legitimate user without having to provide expensive equipment such as CCTV in a small credit card merchant.

In addition, according to the present invention, it is possible to provide a user authentication method and apparatus capable of confirming or authenticating whether a user is a true user in a transaction using a debit card and electronic money.

In addition, according to the present invention, it is possible to provide a user authentication method and apparatus capable of improving a speech recognition rate by adding a natural language processing function, a numeric sound processing function, a keyword detection function, and a noise processing function.

Claims (11)

In the user authentication method, Receiving a voice signal spoken by the user from a terminal; Extracting a speaker component from the received voice signal by performing a preprocessing process; Extracting feature elements from the extracted speaker components; Comparing the extracted feature element with a predefined feature element stored in a storage device; Determining whether the extracted feature element is valid; Generating a message corresponding to the determination result; And Transmitting the generated message to the terminal Including, Using the message to perform at least one of a credit card transaction, a debit card transaction, and an electronic money transaction; User authentication method characterized in that. The method of claim 1, The terminal is Receiving the voice signal through a microphone provided therein User authentication method characterized in that. The method of claim 1, Further comprising: converting the voice signal to a digital signal; User authentication method characterized in that. In the user authentication method, Receiving a speaker component extracted from a voice signal spoken by the user from a terminal; Extracting feature elements from the received speaker component; Comparing the extracted feature element with a predefined feature element stored in a storage device; Determining whether the extracted feature element is valid; Generating a message corresponding to the determination result; And Transmitting the generated message to the terminal Including, Using the message to perform at least one of a credit card transaction, a debit card transaction, and an electronic money transaction; User authentication method characterized in that. The method of claim 4, wherein The terminal is Extracting the speaker component from the speech signal by performing a preprocessing process User authentication method characterized in that. The method of claim 4, wherein The terminal is Receiving the voice signal through a microphone provided therein User authentication method characterized in that. The method of claim 4, wherein The terminal is Analog to Digital Convert the voice signal User authentication method characterized in that. In the authentication device for authenticating a user, Means for receiving a voice signal spoken by the user from a terminal; Means for extracting speaker components from the received speech signal by performing a preprocessing process; Means for extracting feature elements from the extracted speaker components; Means for comparing the extracted feature element with a pre-set feature element stored in a storage device; Means for determining whether the extracted feature element is valid; Means for generating a message corresponding to the determination result; And Means for transmitting the generated message to the terminal Including, Using the message to perform at least one of a credit card transaction, a debit card transaction, and an electronic money transaction; Authentication apparatus for authenticating a user characterized in that. In the authentication device for authenticating a user, Means for receiving a speaker component extracted from a speech signal spoken by the user from a terminal; Means for extracting feature elements from the received speaker component; Means for comparing the extracted feature element with a pre-set feature element stored in a storage device; Means for determining whether the extracted feature element is valid; Means for generating a message corresponding to the determination result; And Means for transmitting the generated message to the terminal Including, Using the message to perform at least one of a credit card transaction, a debit card transaction, and an electronic money transaction; Authentication apparatus for authenticating a user characterized in that. An authentication server for authenticating a user, A memory in which a program is stored; A processor coupled to the memory to execute the program Including, The processor by the program, Receives a voice signal spoken by the user from a terminal, extracts a speaker component from the received voice signal by performing a preprocessing process, extracts a feature element from the extracted speaker component, and stores the extracted feature element Compare with the previously-prepared feature element stored in the terminal, determine whether the extracted feature element is valid, generate a message corresponding to the determination result, and transmit the generated message to the terminal, wherein the terminal uses the message. Thereby performing at least one of a credit card transaction, a debit card transaction, and an electronic money transaction. An authentication server for authenticating a user, A memory in which a program is stored; A processor coupled to the memory to execute the program Including, The processor by the program, Receives a speaker component extracted from the voice signal spoken by the user from the terminal, extracts a feature element from the received speaker component, compares the extracted feature element with a pre-set feature element stored in the storage device, and extracts Determine whether a feature is valid, generate a message corresponding to the determination result, and transmit the generated message to the terminal, wherein the terminal uses the message to perform credit card transactions, debit card transactions, and electronic money transactions; Authentication server for authenticating a user, characterized in that to perform at least one.