JPS63167399A - Voice recognition system - 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 [Object of the Invention] (Field of Industrial Application) The present invention relates to a voice recognition system to be used for account inquiry services by telephone, etc., which are implemented in financial institutions, for example.

(Prior Art) For example, financial institutions have traditionally provided account inquiry services by telephone.

The Kome service uses a voice recognition system that recognizes the voice sent by the user so that the service can be provided not only from a bushphone telephone but also from a rotary telephone.

This voice recognition system first requires the user to say a password that only the user himself or herself knows, then recognizes the voice signal, confirms the user's qualifications, and then executes the service. There is.

(Problem to be Solved by the Invention) According to this method, when a user attempts to receive a service from a public phone, for example, someone nearby is interrupted by a voice indicating the PIN number, and the PIN number is not known. There is a risk of being damaged.

As a result, the service may be used illegally and the privacy of users may be violated.

The present invention was made in view of the above-mentioned circumstances, and it is an object of the present invention to provide a voice recognition system in which there is no fear that a nearby hermit will know the user's password even when the user is attempting to receive a service from a public telephone or the like. It is an object.

[Structure of the Invention] (Means for Solving the Problems) The present invention is characterized in that a transmitting device and a receiving device are connected, and an audio signal corresponding to a multi-digit code number uttered by the transmitting device is received by the transmitting device. In the voice recognition system configured to be recognized by a device, the receiving device is configured to have the receiving device utter a calculation result of a random number that is sequentially created and an actual code number, prior to the uttering of the code number by the transmitting device. A processing means is provided for sending a prompting message to the transmitting device, recognizing the voice signal transmitted in accordance with the prompting message, and converting the recognition result into the actual code number based on the random number value.

(Function) In the voice recognition system of the present invention, the processing means of the receiving device requests the transmitting device to speak not the actual code number, but a code number that is camouflaged by calculation with random numbers that are sequentially created. , Even if your voice is heard by someone nearby, your actual code number will not be known.

゛(Example) Hereinafter, details of an example of the present invention will be described based on the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

In the figure, 1 is a host computer that is equipped with a database that stores the correct PIN numbers of all users in advance and performs the processing necessary for various services; 2 is a host computer that receives voice signals from the user's phone 3 and inputs the PIN numbers as described later. The voice response recognition devices 4a and 4b, which perform voice recognition, are switching devices that have voice signal modulation and demodulation functions and control communication between the voice response recognition device 2 and the user telephone 3.

In this embodiment, when a user requests a service, he or she must first utter the password, but before this utterance, a random number is generated in the voice response recognition device 2, and this random number is compared with the actual number. A message is sent to the user asking them to say the sum of the password and the password.

Here, the voice response recognition device 2 stores the generated random number, calculates the correct password from that value and the recognition result of the number uttered by the user, and sends the obtained password to the user using a synthesized voice. Send it to the other side and have them check whether it is correct or not.

The speech recognition process in this embodiment will be specifically explained below with reference to FIG.

In this example, the correct password for the user is r01.
Suppose that it is 23J.

First, if the voice response recognition device 2 obtains a random number ``5'', it will ask the user ``please add 5 to the first digit (of your PIN number)''. Send tsage.

The value "5" added here is necessary for later conversion, so
Store it in a predetermined memory area.

The user responds to this message by adding “5” to the actual value rOJ.
” and say “5”.

Next, if the random number "3" is 19 in the voice response recognition bag @2, the user will be prompted with "2 (of your PIN)".
Please add 3 to the digit and say the value.''

The value "3" added here is also necessary for later conversion, so
Store it in a predetermined memory area.

The user responds to this message by adding “3” to the actual value “1”.
” and say “4”.

After that, the voice response recognition device 2 uses random numbers "7" and "
If a value of ``8'' is obtained, the user is sent the following messages: ``Please add 7 to the 3rd digit,'' and ``Please add 8 to the 4th digit.'' In response to this message, the user utters "9" and "1".

Note that for the fourth digit, adding "8" to the actual value "3" results in "11", but in such a case, only the first digit of the obtained value should be uttered. Decide in advance.

Thus, according to this embodiment, the voice response recognition device 2 requests the user to say not the actual PIN number, but a number that is camouflaged by calculation with random numbers that are sequentially created. Even when requesting a service from a public telephone or the like, there is no fear that the password will be known to others nearby, and unauthorized use can be prevented and privacy can be protected.

Note that in this embodiment, a random number is added to each digit of the user's personal identification number and is uttered, but the present invention is not limited to this.

In other words, before the user utters the PIN number, all that is required is to prompt the user to utter the result of calculating the random numbers that are created sequentially and the actual PIN number, and the digits to which the random number is added are only some of the digits of the PIN number. It may be. Further, the random number may not only be added to each digit of the PIN number, but also be subtracted from each digit of the PIN number.

Further, the number transmitted by the user is not limited to a personal identification number, but may be any code number, such as an account number or area number, as long as it is a string of numbers that the user does not want a third party to know.

[Effects of the Invention] As explained above, in the voice recognition system of the present invention, the voice recognition system of the present invention requests the utterance of a number that is camouflaged by calculation with random numbers that are sequentially created, rather than the actual PIN number.
Even when a user is trying to receive a service from a public telephone or the like, there is no fear that someone nearby will know his/her PIN number.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, and FIG. 2 is a diagram showing the operation of the system of the embodiment. 1...Host computer 2...Voice response recognition device 3...User telephones 4a, 4b...Switching device

Claims (3)

[Claims] (1) In a voice recognition system configured such that a transmitter and a receiver are connected and the receiver recognizes a voice signal corresponding to a multi-digit code number uttered by the transmitter, the receiver Prior to the transmitting device's utterance of the code number, the device sends a message to the transmitting device prompting the transmitting device to utter the result of the calculation between the random numbers that are sequentially created and the actual code number, and an audio signal is transmitted in accordance with the message. A speech recognition system comprising processing means for recognizing the code number and converting the recognition result into the actual code number based on the random number value. (2) The processing means of the receiving device creates a random number with the same number of digits as the actual code number, and prompts to utter the sum of each digit of this random number and each digit of the actual code number. 2. The speech recognition system of claim 1, wherein the system is configured to send messages to a sending device. (3) Claims 1 or 2, wherein the multi-digit code number is a password for identifying the source of the sound on the transmitting device side.
Speech recognition system as described in section.