JP2003283659A - Method for registering authentication voice data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for registering authentication voice data in which convenience of registering authentication voice data can be enhanced at a server. <P>SOLUTION: The management server 4 in a service management center 3 connected with portable telephones through a portable telephone network 2 stores voice data for recognizing a specified speaker being compared with a voice made by a user and used for personal authentication. Voices of a plurality of users constituting a group have been recorded previously on a voice recording medium, and the voice data for recognizing a specified speaker is formed by reproducing the voices of respective users constituting a group sequentially from the voice recording medium. The voice data for recognizing a specified speaker is associated with the relevant user and stored collectively in the service management center 3. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a registration method of authentication voice data.

[0002]

2. Description of the Related Art In recent years, a mobile phone as a mobile terminal in a digital mobile communication system, PHS (Personal Han
The spread of the dyphone system) continues to grow. These mobile phones and PHS not only communicate with the other party's phone via the mobile phone network, but also exchange various information by accessing servers (computers) of various service centers via the mobile phone network, mobile banking, tickets. The usage is increasing more and more for use such as reservation.

Generally, when a user uses a mobile phone to access a server of a service center registered (contracted) in advance via a mobile phone network, the server requests the mobile phone to input a user name and a password. To do. Then, the server authenticates the user name and password transmitted from the mobile phone, so that the mobile phone is permitted (logged in) to connect to the server. When the connection is permitted, data is exchanged between the mobile phone and the server according to the program for enjoying the service, and the user acquires the desired service.

However, as described above, in the form of personal authentication using the user name and password, anyone can log in if the user name and password are known to others. In other words, security is a problem because another person can impersonate the user of the mobile phone and enjoy the service.

Therefore, as another form of personal authentication, personal authentication using a user's voice print (speech characteristic component) has been put into practical use. Specifically, the user speaks to the mobile phone according to an instruction from the server side when accessing the server of the service center for which a service contract has been made for the first time from the mobile phone. At this time, the server extracts the characteristic component of the user's voice and registers it in the database as authentication voice data. After that, when logging in to the server, the user speaks to the mobile phone according to an instruction from the server side, and the server compares the voice with registered voice data for authentication.
As a result, personal authentication is performed and the connection between the mobile phone and the server is permitted.

[0006]

By the way, in such personal authentication by the voiceprint of the user, when the user accesses the server of the service center from the mobile phone for the first time,
It is necessary to transmit own voice and register authentication voice data for verification based on this voice in the database of the server (initial registration). Therefore, the user has to be bothered for such registration via the mobile phone.

An object of the present invention is to provide a registration method of authentication voice data which can improve the convenience of registration of authentication voice data in a server.

[0008]

In order to solve the above problems, the invention according to claim 1 is stored in a server of a management center to which a mobile terminal is connected via a mobile communication network,
In the registration method of the authentication voice data to be used for personal authentication in comparison with the voice uttered by the user of the mobile terminal,
The voices of a plurality of users that form a group are recorded in advance in a voice recording medium, and the voices of the users that form the group are sequentially reproduced from the voice recording medium to form authentication voice data. The gist is to store the data in the server in association with the user.

According to a second aspect of the present invention, in the authentication voice data registration method according to the first aspect, the authentication voice data is the voice reproduced from the voice recording medium, and the voice quality is equivalent to that of the portable terminal. The main point is to be formed.

The invention described in claim 3 is the invention according to claim 1 or 2.
In the method for registering the authentication voice data described in [1], the authentication voice data is formed by extracting a characteristic component from the voice reproduced from the voice recording medium.

According to a fourth aspect of the present invention, in the authentication voice data registration method according to any one of the first to third aspects, the user is used to associate the user with the authentication voice data. The gist is to carry out based on at least one of the telephone number and the production number of the mobile terminal.

(Operation) According to the invention described in claim 1 or 4, the voices of a plurality of users forming the group are recorded in advance in the voice recording medium, and the voice data for authentication are included in each of the groups. It is formed by sequentially reproducing the voice of each user from the voice recording medium. Then, the authentication voice data is stored in the server of the management center in association with the user. Therefore, it is not necessary for the users in the group who have completed the batch registration to register the authentication voice data even if the user accesses the server of the management center for the first time. That is, the users in the group do not need to register (initial registration) the authentication voice data via the mobile terminal, and only need to record their own voice in the voice recording medium using another recording device, which is convenient. The property is improved.

According to the second aspect of the present invention, the authentication voice data is formed by converting the voice reproduced from the voice recording medium into a sound quality equivalent to that of a mobile terminal. Therefore, the mismatch based on the quality difference between the voice uttered by the user of the mobile terminal and the voice data for authentication is avoided, and the authentication accuracy is improved.

According to the third aspect of the present invention, the authentication voice data is formed by extracting a characteristic component from the voice reproduced from the voice recording medium. Therefore, since the comparison between the voice uttered by the user of the mobile terminal and the voice data for authentication is only the characteristic component, the calculation load is reduced.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram of a communication system using a mobile phone 1. A mobile phone 1 as a mobile terminal is connected to a management server 4 owned by a service management center 3 via a mobile phone network 2 as a mobile communication network. The management server 4 of the service management center 3 is connected via the Internet 5 to a content server 7 of a site (content provider) 6 that provides various contents. Then, the user can enjoy the service owned by the management center 3 by registering the mobile phone 1 to be used in the management center 3 in advance.

FIG. 2 is a block diagram showing an electrical configuration of the management server 4 of the service management center 3. The management server 4 includes an input unit 11, a display unit 12, a communication unit 13, and a storage unit 1.
4, a voice unit 15 and a control circuit unit 16.

The input unit 11 is composed of a keyboard and a mouse (pointing device), and a signal based on an input operation is output to the control circuit unit 16. Input section 1
1 is operated when registering the user's mobile phone 1.

The display unit 12 is composed of a display device and displays various information on its screen based on a control signal from the control circuit unit 16. The communication unit 13 includes the control circuit unit 1
The transmission signal from 6 is output and transmitted to the mobile phone 1 via the mobile phone network 2. In addition, the communication unit 13 uses the mobile phone network 2
The transmission signal from the mobile phone 1 received via the is output to the control circuit unit 16.

The storage unit 14 has a magnetic disk, an optical disk, a magneto-optical disk or the like as a storage medium. The storage unit 14 includes a program database 14a storing various programs for the management server 4 to provide a service to each mobile phone 1, and a specific speaker recognition as authentication voice data registered in a mode described later. The specific speaker voice database 14b storing the voice data for storing is stored. The control circuit unit 16 executes various processes by reading these programs and data.

The audio section 15 uses an AD converter 15a for converting an analog signal obtained by reproducing audio recorded on an audio recording medium (audio tape or the like) 20a by the audio device 20.
It is converted into a digital signal via and output to the control circuit unit 16.

Here, the program database 14
The various programs stored in a include a specific speaker registration program, an authentication program, a service providing program, and the like.

The specific speaker registration program includes an identification number (a mobile phone number, a manufacturing number, etc.) for each user, an individual name,
A screen prompting the registration of voice data for gender, affiliation, and specific speaker recognition is displayed on the display unit 12. Then, the specific speaker registration program stores the identification number, individual name, sex, and affiliation of each user in the specific speaker voice database 14b by the input operation of the input unit 11. In addition, the specific speaker registration program reproduces the voice of each user recorded in advance in the voice recording medium 20a as an analog signal in the audio device 20, so that the sound quality equivalent to that of the mobile phone 1 is obtained via the AD converter 15a. Then, a digital signal whose sampling frequency and number of quantization bits are set so as to drop the signal is generated. For example, the specific speaker registration program has a sampling frequency of 22 kHz so as to reduce the sound quality to that of the mobile phone 1.
z and the number of quantization bits are set to 8 bits. Then, the specific speaker registration program extracts a characteristic signal from the digital signal with the sampling frequency reduced and stores it in the specific speaker voice database 14b as voice data for specific speaker recognition in association with each user. As shown in FIG. 4, the specific speaker voice database 14b stores a table 14c in which identification numbers, individual names, sexes, affiliations, and voice data for specific speaker recognition are associated with each user. .

In the present embodiment, when registering a plurality of users in a group such as an application at a company, the voices of all users are collectively recorded on the voice recording medium 20a and submitted to the service management center 3. It is like this. Therefore, the service management center 3 collectively registers the users in the group by sequentially reproducing the submitted audio recording medium 20a in the above-described manner. In other words, the users in the group who have completed the batch registration do not need to register the voice data for the specific speaker recognition even for the first access to the management server 4 of the service management center 3.

The authentication program recognizes whether the speaker of the voice transmitted from the mobile phone 1 is a speaker registered in advance in the management server 4 (specific speaker recognition) and authenticates it. Specifically, the authentication program extracts the characteristic signal from the voice signal (digital signal) of the talker received from the mobile phone 1 via the communication unit 13. Then, the authentication program stores the voice data (feature signal) for the specific speaker recognition of the user registered in advance corresponding to the mobile phone 1 (mobile phone number, manufacturing number, etc.) in the specific speaker voice database 4
It is read from b and compared with the characteristic signal extracted at this time. Then, the authentication program shifts to a process of providing a service to the user of the mobile phone 1 when the two characteristic signals match. On the other hand, if the two characteristic signals do not match, the authentication program terminates the call being performed with the mobile phone 1.

The service providing program is a program for providing various services to the user of the mobile phone 1. The management server 4 provides various services to each mobile phone 1 according to these programs.

FIG. 3 is a schematic block diagram showing an electrical configuration of the control circuit section 16. The control circuit unit 16 includes a central processing unit (CPU) 21 and an R connected to the CPU 21.
An OM 22, a RAM 23, an input / output interface 24 and a disk controller 25 are provided. The CPU 21, ROM 22, RAM 23, input / output interface 24, and disk controller 25 are connected to the bus 2
They are connected to each other via 6.

The CPU 21 executes various programs by using the data temporarily stored in the RAM 23 and realizes processes necessary for user registration, authentication, and service provision, for example.

The input / output interface 24 includes the input section 11
And an interface for connection with the display unit 12. The disk controller 25 controls driving of the storage unit 14 and input / output of data.

Next, a processing flow of user registration by the management server 4 in this embodiment will be described with reference to FIG. This flow is executed by starting the specific speaker registration program. When this program is started, C
The PU 21 has an identification number for each user, an individual name, a gender,
A screen prompting the registration of voice data for affiliation and specific speaker recognition is displayed on the display unit 12 (step S101). Then, the CPU 21 temporarily stores the identification number, individual name, sex, and affiliation of each user in the RAM 23 by the input operation of the input unit 11 (step S102). Then CPU2
1 inputs the voice of each user recorded in advance in the voice recording medium 20a as an analog signal in the audio device 20 (step S103), and AD
A signal that has been encoded so as to reduce the sound quality to that of the mobile phone 1 is generated via the converter 15a (step S10).
4). Then, the CPU 21 extracts a characteristic signal for recognizing the specific speaker from the signal with the reduced sound quality (step S105), and temporarily stores it in the RAM 23 as voice data for recognizing the specific speaker (step S106). .
In this state, the registration and setting is performed by the input operation of the input unit 11, so that the CPU 21 associates the identification number of each user, the personal name, the sex, the affiliation, and the voice data for the specific speaker recognition with the specific speaker voice. It is stored (registered) in the database 14b (step S107).

Next, a processing flow when the service is provided from the service management center 3 after the user registration is completed in the above-described manner will be described with reference to FIG.

First, the user makes a call from the mobile phone 1 to the service management center 3 (step S201). The mobile phone 1 generates a signal for calling the service management center 3 and transmits it to the service management center 3 via the mobile phone network 2. The management server 4 of the service management center 3 connects to the mobile phone 1 in response to the call signal, and first operates according to the authentication program. That is, the management server 4
The identification number of the mobile phone 1 transmitted following this calling signal is received, and a message requesting voice input for authentication is transmitted to the mobile phone 1 (step S
202). This is because the user registration has been completed in the management server 4 in a lump and there is no need to confirm the user registration.

When the mobile phone 1 receives the message from the management server 4, the mobile phone 1 outputs the message to its display unit or the like (step S203). When the user confirms this message and utters it to the microphone of the mobile phone 1 for authentication in order to respond to this message (step S204), the transmitted voice signal generated based on this utterance is the mobile phone network 2 Is transmitted to the management server 4 via.

The management server 4 generates a characteristic signal (voice data) from the voice uttered by the user based on the received voice transmission signal from the mobile phone 1. Then, the management server 4 performs a matching process with the voice data and the voice data linked with the identification number of the mobile phone 1 stored in the specific speaker voice database 14b (step S20).
5). As a result of the collation, when the voice of the calling user does not match the registered voice, the management server 4 determines that the mobile phone 1
Terminates the call made between and.

On the other hand, as a result of the collation, when the voice of the calling user matches the registered voice, that is, when the voice is authenticated, the management server 4 logs in as the user who has been registered (step S206). That is, when logging in, the management server 4 terminates the authentication program and executes the service providing program. Further, when logging in, the management server 4 sends a notification signal indicating that the user has logged in to the mobile phone 1, and at the same time, a message requesting that a plurality of services to be provided and a desired service to be selected from among the plurality of services are selected. Is transmitted to the mobile phone 1.

When the mobile phone 1 receives the log-in notification signal, it is permitted to access the management server 4 and switches to a processing mode (service processing mode) for receiving various services to the management server 4. That is,
The user can use the mobile phone 1 to operate the management server 4 to receive various services.

Then, upon receiving the message from the management server 4, the mobile phone 1 outputs the message to its display section (step S207). Then, the user operates the operation unit of the mobile phone 1 based on this message to select the service to be received (step S2).
08). The CPU 21 of the management server 4 recognizes the selected service by receiving the signal based on the above operation. Here, for example, when the selected service is “concert ticket reservation”, the management server 4
The content server 7 of the site (content provider) 6 for preserving various concert tickets is accessed via the Internet 5 (step S209).
The management server 4 is provided with a URL (Uniform Resource Locato) for a site (content provider) 6 that provides various services such as reservation sale of concert tickets.
It is registered with r) and is accessed based on the URL.

When the content server 7 is accessed,
Homepage data is transmitted from the content server 7 to the management server 4 (step S210). The homepage data is transmitted to the mobile phone 1 (step S21).
1). That is, the management server 4 functions as a gateway server that manages the mobile phone network 2.

Thereafter, the user operates the operation unit of the mobile phone 1 to exchange various data with the content server 7 via the management server 4. That is, the user of the mobile phone 1 can make a reservation for the desired concert ticket.

As described in detail above, according to this embodiment, the following effects can be obtained. (1) In the present embodiment, the voices of a plurality of users forming a group are recorded in advance in the voice recording medium 20a,
The voice data for recognizing a specific speaker is formed by sequentially reproducing the voices of the users constituting the group from the voice recording medium 20a. Then, the voice data for recognizing the specific speaker is collectively stored in the management server 4 of the service management center 3 in association with the user. Therefore, it is not necessary for the users in the group who have completed the collective registration to register the voice data for recognizing the specific speaker even when the service management center 3 is accessed for the first time. That is, the users in the group are mobile phones 1
There is no need to register (initial registration) the voice data for recognition of a specific speaker via the other way, and it is sufficient to record own voice in the voice recording medium 20a using another recording device, so that convenience can be improved. . Also, the communication cost associated with the initial registration via the mobile phone 1 can be reduced.

(2) In this embodiment, the voice data for recognizing a specific speaker is formed by converting the voice reproduced from the voice recording medium 20a into the sound quality equivalent to that of the mobile phone 1. Therefore, it is possible to avoid the inconsistency based on the quality difference between the voice uttered by the user of the mobile phone 1 and the voice data for the specific speaker recognition, and the authentication accuracy can be improved.

(3) In the present embodiment, the voice data for specific speaker recognition is formed by extracting a characteristic signal from the voice reproduced from the voice recording medium 20a. Therefore,
Since the comparison between the voice uttered by the user of the mobile phone 1 and the voice data for recognizing the specific speaker is only the characteristic signal, the calculation load can be reduced.

The embodiment of the present invention is not limited to the above embodiment, but may be modified as follows. In the above-described embodiment, the voice of the user is analog-recorded (recorded) on the voice recording medium (audio tape or the like) 20a. On the other hand, the user's voice may be digitally recorded (recorded) on, for example, an MD (mini disk) or a DAT (digital audio tape). Then, in this case, the digitally recorded user's voice may be reproduced as a digital signal to form the specific speaker voice data for the specific speaker recognition registered in the management server 4 based on this. If the sampling frequency and the number of quantization bits of the digitally recorded user's voice are higher than the sound quality of the mobile phone, the sampling frequency and the number of quantization bits can be reduced to the sound quality equivalent to that of the mobile phone 1. The same effect as that of the above embodiment can be obtained.

In the above-described embodiment, cue information or tracking information may be recorded between the voices of the users in addition to recording the voices of a plurality of users constituting the group on the voice recording medium 20a. . Then, when the voice of each user forming the group is reproduced from the voice recording medium 20a, the reproduction / pause is automatically performed by using the recording information, and the voice data for the specific speaker recognition is sequentially formed.・ You may remember.

In the above embodiment, the voice section 15 is built in the management server 4, but a similar voice section may be provided externally. In the above embodiment, the mobile phone 1 is used as the mobile terminal.
PHS (Personal Handyphone Syste
m) or PDA (Personal Digital Assistant).

In the above-described embodiment, the service related to authentication is provided as a membership service (content,
Applications, etc.), electronic commerce, companies providing highly confidential information, and e-government providing administrative services (for example, application for resident card, seal certification, etc.).

Next, the technical idea which can be understood from the above embodiment will be described below together with its effects. (A) A server of a management center to which a mobile terminal is connected via a mobile communication network, which stores authentication voice data which is compared with a voice uttered by a user of the mobile terminal and is used for personal authentication. In the above, the authentication voice data formed by sequentially reproducing the voice of each user from the voice recording medium in which the voices of a plurality of users constituting the group are recorded in advance are stored in association with the user. And the server. According to this configuration, the voices of the plurality of users that form the group are recorded in advance on the voice recording medium, and the voice data for authentication reproduces the voice of each user that forms the group sequentially from the voice recording medium. Formed. Then, the authentication voice data is stored in the server of the management center in association with the user.
Therefore, it is not necessary for the users in the group who have completed the batch registration to register the authentication voice data even if the user accesses the server of the management center for the first time. That is, the users in the group do not need to register (initial registration) the authentication voice data via the mobile terminal, and only need to record their own voice in the voice recording medium using another recording device, which is convenient. The property is improved.

[0047]

As described above in detail, according to the invention described in claims 1 to 4, the convenience of registration of the authentication voice data in the server can be improved.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a mobile communication system in the present embodiment.

FIG. 2 is a block diagram showing an electrical configuration of a server.

FIG. 3 is a block diagram showing an electrical configuration of a control circuit unit.

FIG. 4 is an explanatory diagram showing a table of a specific speaker voice database.

FIG. 5 is a flowchart showing a processing mode of user registration by the management server 4.

FIG. 6 is a flowchart showing an aspect of providing a service to a mobile phone.

[Explanation of symbols]

1 Mobile phone as a mobile terminal 2 Mobile phone network as mobile communication network 3 Service management center Management server as 4 servers 20a audio recording medium

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G10L 15/00 G10L 3/00 521P 15/06 521C 17/00 551A // H04L 9/32 545D H04L 9 / 00 673DF Term (reference) 5B085 AE27 BA06 BG02 BG03 5D015 AA03 GG03 KK02 5J104 AA07 KA01 KA16 KA18 PA02 5K024 AA62 BB01 BB05 EE09 GG05

Claims (4)

[Claims] 1. Registration of authentication voice data which is stored in a server of a management center to which a mobile terminal is connected via a mobile communication network and which is used for personal authentication by being compared with a voice uttered by a user of the mobile terminal. In the method, voices of a plurality of users constituting a group are recorded in advance in a voice recording medium, voices of respective users constituting the group are sequentially reproduced from the voice recording medium to form authentication voice data, A method for registering authentication voice data, characterized in that the authentication voice data is stored in the server in association with the user. 2. The registration of the authentication voice data according to claim 1, wherein the authentication voice data is formed so that the voice reproduced from the voice recording medium has a sound quality equivalent to that of the mobile terminal. Method. 3. The registration of the authentication voice data according to claim 1, wherein the authentication voice data is formed by extracting a characteristic component from a voice reproduced from the voice recording medium. Method. 4. The association with the authentication voice data of the user is performed based on at least one of a telephone number and a manufacturing number of a mobile terminal used by the user. How to register the voice data for authentication described in Crab.