JPS6294899A - Personal identification system using ic card - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] (fl) Technical field This invention relates to a personal authentication system using an IC card, and in particular, a personal authentication system that extracts the characteristics of the spoken language of a card user so that it can be used only by the user himself/herself. Regarding.

(b) Prior Art and Its Disadvantages Conventional personal authentication systems use magnetic cards with data recorded on magnetic stripes to authenticate individuals, or systems in which a PIN number set by the user is registered in advance. This PIN is now being used for personal authentication, or a combination of these. Such conventional personal authentication systems have a risk that, for example, if the magnetic card falls into the hands of another person, it may be used by someone other than the user. Furthermore, since the password is usually a number that is easy to remember, such as a date of birth or a telephone number, there is a problem that it can be easily known. Furthermore, these days, people often carry multiple cards, and there is also the drawback that it is easy to forget or misremember the PIN number.

(C1 Purpose of the Invention The purpose of this invention is to create a personal authentication system that can reliably perform personal authentication without being used by anyone other than the individual, even if the card falls into the hands of another person. Our goal is to provide the following.

((1) Structure and Effects of the Invention The present invention provides an IC card capable of storing speaker voice characteristic data. a first speaker feature data extracting means for extracting speaker feature data from voice data input from the speech input section of the speech input section; a voice registration device having a speaker characteristic data writing means for writing speaker characteristic data into the storage section of the IC card as person characteristic data; a second voice input section for inputting voice; and a voice input from the second voice input section; a second speaker feature data extracting means for extracting speaker feature data of the data, and the speaker feature data extracted by the second speaker feature data extracting means and the reference speaker feature read from the IC card. The present invention is characterized in that it is comprised of an authentication device having an authentication means for performing speaker authentication by comparing data.

In general, the frequency characteristics of speech signals of spoken words are determined by the shape of the vocal tract including the lips, tongue, jaw, and the like. The shape of such a vocal tract is individual, and therefore speaker characteristic data that identifies the speaker can be extracted from the audio signal of spoken words.

By configuring as described above, the speaker's characteristic data contained in the person's spoken words that express the person's individuality can be
It is stored in the card, and when necessary, the speaker characteristic data included in the spoken words of the user of the IC card and the IC
It is possible to authenticate the identity of the user of the IC card by comparing it with the speaker characteristic data written on the card. Therefore, even if the IC card falls into the hands of another person, the speaker characteristic data included in that person's spoken words will not match the speaker characteristic data stored in the (0 car F), so the It is possible to identify that the user is not the user himself/herself.Also, since the method does not rely on the user's memory using a PIN number, there are no problems caused by forgetting or misremembering the PIN number.

TELE EMBODIMENT FIG. 1 is a diagram showing the configuration of a voice registration device of a personal authentication system which is an embodiment of the present invention. 11 is a microphone; 12 is a voice analysis device that extracts speaker characteristic data from the voice signal input from the microphone; 13 is a controller that controls the entire voice registration device; It temporarily stores speaker characteristic data and outputs speaker characteristic data to be written to the tC card-like 14 as necessary. The IC card reader 14 writes speaker characteristic data to the IC card inserted through the IC insertion slot 4a. The IC card 5 is provided with a nonvolatile memory and an IC chip of a CPU that reads and writes data to the memory. This IC
The speaker characteristic data and the like once written in the voice registration device 5 retain their stored contents until they are written again by this voice registration device.

FIG. 2 shows a gate terminal device including an authentication device. 2 is a main unit, 21 is an IC card insertion slot, 22 is a microphone that receives the user's voice, 23 is a talk switch that is operated when inputting voice, 24 is a display that informs the user of the timing of voice input, 25 26 is a display that lights up when the authentication result is established, and 26 is a display that lights up when the authentication result is not established.

The user inserts the IC card on which the user's speaker characteristic data is written using the voice registration device shown in FIG. 1 into the IC card insertion slot 21 shown in FIG. 2, and!・-Kuswitch 2
By pressing 3 and inputting a predetermined word, such as your name, by voice, you can pass through the gate. Even if someone who is not an IC card user uses this gate terminal device, authentication will not be established and they will not be able to pass through the gate.

FIG. 3 is an overall block diagram of the voice registration device.

The audio signal input from the microphone 11 is sent to the amplifier circuit 1
21, and a low-pass filter 122 cuts off a frequency band higher than the sampling frequency.

123 is a sample hold circuit which samples the audio signal in synchronization with the sampling frequency. A/D
Converter 124 converts the value sampled by 123 into a digital value and outputs it to analysis circuit 125 .

The analysis circuit 125 is a parallel/serial conversion circuit for converting the digitized audio signal, and outputs the parameters determined by the analysis circuit 125 as serial data. These blocks 121 to 126 constitute the speech analysis device (see FIG. 12), and correspond to the first speaker characteristic data extraction means of the present invention.

The CPUI 30 is composed of a microprocessor,
The operation is performed by program processing. This control program is recorded in the ROM 135 in advance. RAM136
is used as a working area when executing this control program, and is used as a working area for the parallel/serial conversion circuit 12.
The parameters output from 6 are temporarily stored. l
I1 10131 reads data output from the parallel/serial conversion circuit 126 and outputs data to be written to the IC card 5 to the IC card writer 14.
0 boats. Reference numeral 132 is an I10 port for reading the contents of the keys on the keyboard 133 manually. A display memory 134 stores data to be displayed on the display unit 138, and a display control circuit 137 performs display control based on the contents of this memory. Note that 130 to 138 constitute the controller (see 13 in FIG. 1).

FIG. 6 is a flowchart showing the processing procedure of the CPUI 30. First, a display requesting the user's ID number is displayed, and a key is read. It determines that the operated key is a numeric key and places it on the cover sofa (n 14-”n 16).N is the number of digits of the ID number.
Once the number of digits has been entered, the ID number is checked against the already registered ID number (n18-n20). The input and verification of this ID number is a process for determining whether the registrant is a registrant who can use the voice registration device shown in FIG.

If the ID numbers match, a display requesting operation of the analysis start key is displayed and the key is read (n24-n26).
. The operator presses the analysis start key and at the same time inputs a keyword into the microphone for about a few seconds. The device determines that the analysis start key has been operated, performs PARCOR analysis on the audio input signal, and temporarily stores the parameters in a predetermined area of the RAM (n30-=n32). After a certain period of time has passed, the analysis process is finished, a display requesting the operation of the authentication check key is displayed, and the key is read (n
36→n38). The operator presses the authentication check key and inputs the same keyword again by voice. The device determines that the authentication check key has been operated, and issues PARc in the same manner as above.

Perform R analysis. As a result, the parameters stored in n32 are compared with the parameters resulting from the current analysis, and it is determined whether they are within a predetermined similarity range. If both parameters are deemed to match, the already memorized parameters are output to the IC card writer (n44→
n46). As a result, the IC card writer writes the reference speaker characteristic data to the IC card.

If the voice input while pressing the authentication check key does not match the previous analysis result, the process returns to step n24 and starts over from the beginning. In this way, the operator can stably pronounce the keyword to be input by voice, and can be trained to perform authentication reliably.

FIG. 5 represents a memory map within the IC card.

The ID number and name are the ID number and name of the user of this IC card, and are information registered by the voice registration device shown in FIG. 1 or other dedicated registration device. Kl-Kn is n analyzed by the speech analysis device.
This parameter is written in the process of step n4G in FIG. 6.

FIG. 4 is a block diagram of a control section of the gate terminal device shown in FIG. 2. Microphone 22. Amplification circuit 201. Low pass filter 202. Sample hold circuit 203.
A/D converter 204° analysis circuit 205. parallel/
The serial conversion circuit 206 is a voice analysis section that performs the same operation as the voice analysis device in the voice registration device shown in FIG. 3, and corresponds to the second speaker characteristic data extraction means of the present invention. The CPU 200 is a microprocessor that controls the entire authentication device. This control program is stored in the ROM 211 in advance. RAM21
2 is used as a working area when executing this program, and temporarily stores, for example, parameters that are analysis results.

The I10 port 207 is a port for reading parameters that are reference speaker characteristic data read from the memory of the IC card 5 by the rc card reader 210, and for reading parameters output from the parallel/serial conversion circuit 206. 208 is an I10 port for reading the state of the talk switch 23; 209 is the display body 2;
It is an I10 boat that turns on/off lights at 4, 25, and 26. The authentication device in the gate terminal device is configured as described above.

214 is a drive circuit that drives the rotating cam of the locking device, and is connected to the CPU locker 1 via the I10 boat 213.
First, the sensor installed in the IC card insertion slot
If it detects the insertion of a C card, it will accept the IC card and
Read out the memory contents (n50-+n52→n54)
o Determine whether the ID number read from the IC card is a usable ID number, and if the number is registered, light up the display 24 indicating that voice input is possible (n 5 G- "n 5 B). The operator confirms that the display 24 is lit, presses the talk switch 23, and inputs the keyword by voice. The device indicates that the talk switch has been activated by '!-11, While starting analysis of the input audio signal, step n64
A comparison is made with the parameters read out from the IC card. If the result is within a predetermined similarity range, the display 25 is lit to indicate that authentication has been achieved (
n 6' 4-=n 6' 6). After that, the IC card is ejected from the insertion slot, it is determined whether or not the IC card has been removed, and if it has been removed, the lock is unlocked (n74-=
n7G). After that, the lock is closed after a certain period of time has passed (n
78-"n80). In this way, the user of the IC card can pass through the gate by inputting the keyword by voice.

If the inserted IC card is an unusable card or if authentication is unsuccessful even if it is a usable IC card, the display 26 indicating this is lit and the IC card is ejected (n68-" n70).

As described above, a personal authentication system using an IC card can be configured.

In the embodiment, speaker characteristic data is extracted and used by PARCOR analysis, but it is also possible to extract and use other data such as pitch, short-time spectrum, cepstrum, linear prediction coefficient, etc.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the configuration of the voice registration device in the personal authentication system of the present invention, FIG. 2 is a diagram showing the appearance of a gate terminal device using the authentication device of the same system, and FIG. 3 is a diagram showing the configuration of the voice registration device 4 is a block diagram of the authentication device, FIG. 5 is a diagram showing a memory map of the memory in the IC card, FIG. 6 is a flowchart showing the processing procedure of the CPU of the voice registration device, and FIG. 7 is a block diagram of the authentication device. CP of the authentication device
It is a flowchart showing the processing procedure of U. 2-Game [・Terminal device 51C card, 11-Microphone, 12-Speech analyzer, 13-Controller, 14-IC
Card writer, 2l-IC card insertion slot, 22-Microbon, 23-1--Kusisochi, 24-
26-Display body.

Claims (1)

[Claims] (1) An IC card having a storage section capable of storing speaker voice characteristic data, a first voice input section for inputting voice, and a speaker of voice data input from the first voice input section. a first speaker feature data extracting means for extracting feature data; and writing the speaker feature data extracted by the first speaker feature data extracting means into the storage section of the IC card as reference speaker feature data. A voice registration device having a speaker characteristic data writing means, a second voice input section for inputting voice, and a second story for extracting speaker characteristic data from voice data input from the second voice input section. person characteristic data extraction means, and the second
an IC card comprising an authentication device that performs speaker authentication by comparing the speaker characteristic data extracted by the speaker characteristic data extraction means and the reference speaker characteristic data read from the IC card; A personal authentication system using