JP7553652B2 - User authentication method - Google Patents

Description

The present invention relates to a user authentication method.

In information processing devices and network systems, user authentication is performed by voice input. In user authentication by voice input, a password voice is registered in advance. If the registered password voice matches the password voice entered by the person to be authenticated, the person is authenticated as the registered person.

For example, Patent Document 1 discloses an identity authentication method in which a password voice is registered in combination with a key word for that password voice, and the key word is presented to the user during authentication, and the user is asked to speak the password voice corresponding to the key word. This makes it difficult for others to falsely authenticate.

Japanese Patent Application Publication No. 10-173644

However, in the method disclosed in Patent Document 1, the user must accurately memorize the registered PIN voice and speak it when authenticating, which places a heavy burden on the user. Furthermore, if the user forgets the PIN voice, the user cannot be authenticated and is unable to use the information processing device, etc.

The present invention aims to provide an information processing device, a user authentication network system, and a user authentication method that are capable of performing user authentication using voice input in an easy-to-use and highly secure manner.

The following is a brief summary of the representative inventions disclosed in this application.

An information processing device according to a representative embodiment of the present invention includes a voice input processing unit, a memory, and a controller. In a user registration process for registering a user in the information processing device, the voice input processing unit captures a registration voice uttered by the user to be registered and converts the registration voice into a first voice signal. The controller extracts a first voice characteristic from the first voice signal and compares the first voice characteristic with the voice characteristics of a previously registered user that are already stored in memory. If the first voice characteristic does not match the voice characteristic of the previously registered user, the controller executes a process of storing the first voice characteristic in memory in association with the user to be registered, generating a voice password based on the first voice characteristic, and storing the voice password in the memory. The controller performs a user authentication process for the user to be authenticated using the voice password stored in memory by the user registration process.

The effects achieved by the representative inventions disclosed in this application can be briefly explained as follows:

In other words, according to a representative embodiment of the present invention, it becomes possible to perform user authentication using voice input in an easy-to-use and highly secure manner.

1 is a block diagram showing an example of a configuration of an information processing device according to a first embodiment of the present invention; FIG. 11 is a diagram showing an example of an operation sequence of a user registration process according to the first embodiment of the present invention. FIG. 4 is a diagram showing an example of an operation sequence of a user authentication process according to the first embodiment of the present invention. FIG. 4 is a diagram showing an example of an operation sequence of a user authentication process according to the first embodiment of the present invention. FIG. 11 is a block diagram showing an example of a configuration of a user authentication network system according to a second embodiment of the present invention. FIG. 11 is a diagram showing an example of an operation sequence of a user registration process according to the second embodiment of the present invention. FIG. 11 is a diagram showing an example of an operation sequence of a user authentication process according to the second embodiment of the present invention.

The best mode for carrying out the present invention will be described in detail below with reference to the drawings. In each drawing for explaining the best mode for carrying out the present invention, the same reference numerals are used for components having the same functions, and repeated explanations will be omitted.

(Embodiment 1)
Fig. 1 is a block diagram showing an example of the configuration of an information processing device according to the first embodiment of the present invention. As shown in Fig. 1, the information processing device 100 includes a controller 101, a memory 114, a voice input processing unit 120, a voice output processing unit 121, a display unit 128, a communication unit 129, and a base station communication unit 130. These components constituting the information processing device 100 are connected to each other via a bus 140.

The memory 114 is a functional block that stores various information handled by the information processing device 100. The memory 114 has a non-volatile memory such as a flash memory, and stores various programs executed by the controller 101, and various information such as the voice characteristics and voice passwords of each registered user. The memory 114 includes a program storage area 113 and a user information storage area 106 (e.g., 106a to 106c).

The program memory area 113 is a memory area that stores various programs such as an OS (Operating System) and operation control applications. These programs are executed by the controller 101.

The user information storage area 106a is a storage area that stores user information for, for example, user A, who is a registered user. The user information storage area 106a includes a voice feature storage area 102a that stores the voice features of user A, and voice password storage areas (e.g., 103a to 105a) that store the voice password of user A. The voice password storage areas 103a to 105a each store a different voice password for user authentication of user A. Note that the number of voice password storage areas may be less than this, or may be more than this.

The user information storage area 106b is a storage area that stores user information of, for example, user B, who is a registered user. The user information storage area 106b includes a voice feature storage area 102b that stores the voice features of user B, and voice password storage areas (e.g., 103b to 105b) that store the voice password of user B. The voice password storage areas 103b to 105b each store a different voice password for user authentication of user B. Note that the number of voice password storage areas in the user information storage area 106b may be greater or less than this number.

The user information storage area 106c is a storage area that stores user information of, for example, user C, who is a registered user. The user information storage area 106c includes a voice feature storage area 102c (not shown) that stores the voice features of user C, and voice password storage areas (e.g., 103c to 105c, not shown) that store the voice password of user C. The voice password storage areas 103c to 105c each store a different voice password for user authentication of user C. Note that the number of voice password storage areas in the user information storage area 106c may be greater or less than this.

The audio input processing unit 120 includes a microphone for audio input. The audio input processing unit 120 captures external audio via the microphone and converts the captured audio into an audio signal. The audio input processing unit 120 outputs the audio signal to the memory 114 or the controller 101. The memory 114 may store the audio signal output from the audio input processing unit 120.

The audio output processing unit 121 includes a speaker for audio output. The audio output processing unit 121 outputs audio based on the audio signal to the outside via the speaker.

The controller 101 is a functional block that controls each part constituting the information processing device 100. The controller 101 includes a CPU (Central Processing Unit) and a RAM (Random Access Memory). The controller 101 reads various programs such as an OS and an operation control application from the program storage area 113 of the memory 114 and loads them in the RAM. The controller 101 executes the programs loaded in the RAM with the CPU, thereby realizing each functional block such as the voice feature extraction processing unit 122, the voice feature matching and discrimination processing unit 123, the voice password generation processing unit 124, the voice password extraction processing unit 125, and the voice password matching and discrimination processing unit 126 shown in FIG. 1 in the CPU. The controller 101 controls the operation of the information processing device 100 by using these functional blocks. Note that each functional block in the controller 101 may be realized by software only, or may be realized by software and hardware.

The voice feature extraction processing unit 122 extracts the voice features of the person who spoke the voice from the voice signal output from the voice input processing unit 120. The voice feature extraction processing may utilize well-known technology that is generally used in so-called voice recognition processing. Specifically, the voice signal is frequency analyzed for each section of several tens of milliseconds to obtain frequency information of the voice (acoustic spectrum) in a time series, and this acoustic spectrum is extracted as the voice feature.

The voice feature matching and discrimination processing unit 123 compares the voice features extracted by the voice feature extraction processing unit 122 with the voice features of registered users stored in the voice feature storage area (e.g., 102a to 102c) of the memory 114, and determines whether the two match.

The audio password generation processing unit 124 generates multiple audio passwords based on the audio features stored in the audio feature storage area. For example, the audio password generation processing unit 124 generates multiple audio passwords for user A based on the audio features stored in the audio feature storage area 102a. Similarly, the audio password generation processing unit 124 generates multiple audio passwords for users B and C based on the audio features stored in the audio feature storage areas 102b and 102c, respectively.

In addition, while the information processing device 100 is in use, the voice characteristics of a registered user are updated from time to time. Therefore, the voice password generation processing unit 124 may generate a new voice password based on the updated voice characteristics. The voice password generation processing unit 124 stores the generated voice password in the voice password storage area of the corresponding registered user.

The audio password extraction processing unit 125 extracts an audio password from the voice uttered by the user to be authenticated during user authentication processing.

The audio password matching and discrimination processing unit 126 compares the audio password extracted by the audio password extraction processing unit 125 with the audio password stored in the user information storage area of the memory 114, and determines whether the two match.

The display unit 128 has a display panel such as a liquid crystal panel, and displays various information to be notified to the user. For example, the display unit 128 displays information such as the text of the voice password to be spoken by the user to be authenticated during user authentication processing. The display unit 128 is provided with an input operation unit 127. The input operation unit 127 is an input device such as a keyboard or mouse, and various information is input by the user's operation.

The display panel may be configured as a capacitive touch panel or the like, and may be used as the input operation unit 127. In this case, the user inputs various pieces of information by approaching or touching the touch panel with a finger, a touch pen, or the like.

The communication unit 129 is a communication interface that is connected to an external network, for example, via a wireless LAN or a wired LAN, and transmits and receives information between the external network and devices connected to the external network.

The base station communication unit 130 is a communication interface that performs long-distance wireless communication, such as W-CDMA (Wideband Code Division Multiple Access) or GSM (Global System for Mobile communications). The base station communication unit 130 transmits and receives information between devices connected to the long-distance wireless communication via a transmitting and receiving antenna.

When the information processing device 100 needs to connect to an external database during the voice feature extraction process, the information processing device 100 accesses the external database via the communication unit 129 or the base station communication unit 130, and various information is transmitted and received between the external database and the information processing device 100.

<User registration process>
Next, the user registration process according to this embodiment will be described. Fig. 2 is a diagram showing an example of an operation sequence of the user registration process according to the first embodiment of the present invention. In Fig. 2, the operation of a user to be registered is shown on the left side, and the operation of the information processing device 100 is shown on the right side. The user registration process is a process for registering a user in the information processing device 100. In the user registration process, the following steps S201 to S215 are performed.

First, in step S201, the user to be registered performs an input operation to start the user registration process from the input operation unit 127. For example, the user to be registered operates a registration start button, touches a registration start button displayed on a display panel, or inputs a temporary password.

The information processing device 100 receives a registration start input from the input operation unit 127 (S202) and requests the user to be registered to speak some kind of voice for registration (S203). Specifically, the information processing unit 100 requests the user to be registered by displaying the request for voice speech on the display unit 128 or by transmitting the request by voice from the voice output processing unit 121 via the speaker.

When the user to be registered recognizes the request for speech from the information processing device 100 (S204), the user speaks some kind of voice for registration toward the information processing device 100 (S205).

When the user to be registered speaks a registration voice, the voice input processing unit 120 captures the registration voice, converts it into a voice signal (first voice signal), and outputs the voice signal to the controller 101 via the bus 140 (S206).

The voice feature extraction processing unit 122 extracts the voice feature (first voice feature) of the user to be registered from the voice signal output from the voice input processing unit 120 (S207). The voice feature matching and discrimination processing unit 123 then compares the voice feature of the user to be registered with the voice features of previously registered users that are already stored in the voice feature storage area (e.g., 102a to 102c) of the memory 114 (S208), and determines whether the voice feature of the user to be registered has already been stored in the voice feature storage area (e.g., 102a to 102c) (S209).

If it is determined that the voice features of the user to be registered do not match those of a previously registered user stored in a voice feature storage area (e.g., 102a to 102c) (N), the controller 101 determines that the user to be registered is a new user who has not yet been registered, and stores the voice features of the user to be registered in memory 114 in association with the user to be registered (S210). Specifically, the controller 101 allocates a new user information storage area 106 for this user to be registered, and stores the voice features of the user to be registered in the voice feature storage area 102 of the newly allocated user information storage area 106 (S210).

Then, the voice password generation processing unit 124 generates multiple voice passwords for the user to be registered based on the voice characteristics of the user to be registered, and stores the multiple generated voice passwords in the newly assigned voice password storage area (e.g., 103 to 105) of the user information storage area 106 (S211).

The controller 101 notifies the user to be registered that the user registration process is complete by displaying on the display unit 128 or emitting a sound from the audio output processing unit 121 (step S212). Note that the controller 101 may notify the user to be registered that the user registration process is complete by using a combination of displaying on the display unit 128 and emitting a sound from the audio output display 121.

The user to be registered recognizes that his/her user registration process has been completed by receiving a notification from the information processing device 100 (S213). This completes a series of steps related to the user registration process.

On the other hand, in step S209, if the voice features of the user to be registered match those stored in any of the voice feature storage areas (e.g., 102a to 102c) (Y), the controller 101 determines that the user to be registered is a previously registered user, and notifies the user to be registered that the user has already been registered by displaying on the display unit 128 or emitting sound from the audio output processing unit 121 (step S214). Note that even in this case, the controller 101 may use both the display on the display unit 128 and sound from the audio output display 121.

Then, the user recognizes from the notification from the information processing device 100 that he or she has already been registered as a user (S213). This also ends the series of processes related to user registration.

The information processing device 100 performs these user registration processes each time a request is received from a user to be registered.

<User authentication process>
Next, the user authentication process in this embodiment will be described. Figures 3 and 4 are diagrams showing an example of an operation sequence of the user authentication process according to the first embodiment of the present invention. In Figures 3 and 4 as well, the operation of the user to be authenticated is shown on the left side, and the operation of the information processing device 100 is shown on the right side. Note that Figure 4 shows the detailed processing contents of the main sequence S350 in Figure 3. In the main sequence S350 in Figure 3, only steps showing the connection relationship between the main sequence and other steps outside the main sequence are shown for convenience.

First, in step S301, the user to be authenticated speaks some kind of authentication voice toward the information processing device 100 to start the user authentication process.

The voice input processing unit 121 captures the authentication voice uttered by the user to be authenticated, converts the authentication voice into a voice signal (second voice signal), and outputs the voice signal to the controller 101 via the bus 140 (S302).

The voice feature (second voice feature) of the user to be authenticated is extracted from the voice signal output from the voice input processing unit 120 (S303). The voice feature matching and discrimination processing unit 123 then compares the voice feature of the user to be authenticated with the voice features of registered users already stored in the voice feature storage area (e.g., 102a to 102c) of the memory 114 (S304), and determines whether the voice feature of the user to be authenticated has already been stored in the voice feature storage area (e.g., 102a to 102c) (S305).

If the voice characteristics of the user to be authenticated do not match the voice characteristics of the registered user stored in the voice characteristics storage area (102a to 102c, ...) (N), the controller 101 determines that the user to be authenticated is unregistered and not a registered user, and notifies the user to be authenticated that user authentication processing is impossible by displaying a message on the display unit 128 or emitting a sound from the audio output processing unit 121 (S324).

When the user to be authenticated recognizes from the information processing device 100 that the user authentication process is not possible (S325), the user determines whether or not to perform the user authentication process again (S326). If the user to be authenticated determines that the user authentication process should be performed again (Y), the user speaks the authentication voice again in step S301. If the user to be authenticated determines that the user authentication process should not be performed again (N), the user authentication process ends.

On the other hand, in step S305, if the voice feature matching and discrimination processing unit 123 determines that the voice features of the user to be authenticated match those stored in any of the voice feature storage areas (e.g., 102a) (Y), the controller 101 narrows down the list of specific users to be authenticated by voice password to those already registered users (e.g., user A) whose voice features match those of the user to be authenticated (S306).

When a specific user (e.g., user A) is narrowed down, the controller 101 randomly selects one voice password from the multiple voice passwords stored in the voice password storage area (e.g., 103a to 105a) of the user information storage area (e.g., 106a) corresponding to the specific user (S307). The controller 101 then presents the selected voice password to the user to be authenticated by displaying it on the display unit 128 or emitting sound from the voice output processing unit 121, and prompts the user to be authenticated to speak the presented voice password (S308).

When the user to be authenticated recognizes the voice password notified by the information processing device 100 (S309), he/she speaks the recognized voice password (S310).

After presenting the voice password to the user to be authenticated, the voice input processing unit 120 captures the voice of the voice password spoken by the user to be authenticated, converts the voice of the voice password into a voice signal (third voice signal), and outputs the voice signal to the controller 101 via the bus 140. The voice password extraction processing unit 125 then extracts the voice password spoken by the user to be authenticated from the captured voice signal (S311).

In parallel with these processes, after notifying the user to be authenticated of the audio password, the controller 101 determines whether or not it has captured the audio of the audio password uttered by the user to be authenticated (S312). Furthermore, the controller 101 measures the presentation time that has elapsed since notifying the user to be authenticated of the audio password, and determines whether or not the presentation time has elapsed a preset timeout time (S313).

For example, if the controller 101 determines in step S312 that the voice of the voice password uttered by the user to be authenticated has not been captured (N) and determines in step S313 that the presentation time has not elapsed the timeout time (N), it performs the process of step S312 again. That is, in this case, the information processing device 100 continues to wait so that the voice of the voice password uttered by the user to be authenticated can be captured.

Furthermore, if the controller 101 determines in step S312 that the audio password uttered by the user to be authenticated has not been captured (N) and determines in step S313 that the presentation time has exceeded the timeout time (Y), the controller 101 stops the user authentication process due to the timeout. The controller 101 then notifies the user to be authenticated that the user authentication process has been stopped due to the timeout by displaying a message on the display unit 128 or emitting a sound from the audio output processing unit 121 (S327).

The user to be authenticated recognizes that the user authentication process has been stopped due to a timeout (S328) and determines whether or not to perform the user authentication process again (S326). Step S326 has already been described, so the description of the subsequent process will be omitted here.

On the other hand, if the controller 101 determines in step S312 that it has captured the voice uttered by the user to be authenticated (Y), i.e., that it has captured the voice password within the timeout period, the voice password matching and determination processing unit 126 compares the voice password presented to the user to be authenticated with the voice password extracted from the voice signal (third voice signal) of the user to be authenticated (S314), and determines whether these voice passwords match (S315).

If in step S315 the audio password matching and determination processing unit 126 determines that these audio passwords do not match (N), the controller 101 notifies the authenticated user that these audio passwords do not match by displaying a message on the display unit 128 or by emitting a sound from the audio output processing unit 121 (S329).

The user to be authenticated recognizes the mismatch between these voice passwords through notification from the information processing device 100. That is, the user to be authenticated recognizes that the presented voice password does not match the voice password spoken by the user (S330), and determines whether to speak the voice password again (S331).

If it is determined in step S331 that the user to be authenticated will not re-speak the audio password (N), the process of step S326 is performed. Step S326 has already been described, so its explanation is omitted.

In response to this, if the user to be authenticated determines in step S331 that the voice password is to be re-spoken (Y), the user determines whether or not to request the information processing device 100 to re-notify the voice password (S332). If the user to be authenticated determines in step S332 that the user does not wish to request re-notification of the voice password (N), the user returns to step S310 and re-speaks the same voice password that was previously notified.

On the other hand, if the user to be authenticated determines in step S332 that he or she wishes to request re-notification of the audio password (Y), the user requests re-notification of the audio password from the information processing device 100 by vocalizing or inputting the information (S333).

Then, the information processing device 100 receives a re-notification request from the user to be authenticated (S334) and notifies the user to be authenticated of another audio password for the specific user by performing the processes of steps S307 to S308 again.

Meanwhile, in step S315, if the voice password matching and discrimination processing unit 126 determines that the voice password of the specific user presented to the user to be authenticated matches the voice password extracted from the voice signal (third voice signal) of the user to be authenticated (Y), the voice feature extraction processing unit 122 extracts a voice feature (third voice feature) from the voice signal of the voice password spoken by the user to be authenticated captured in step S311 (S316). Then, the voice feature matching and discrimination processing unit 123 compares and collates the extracted voice feature with the voice feature of the specific user (e.g. user A) (S317) and determines whether the two voice features match (S318).

If the voice feature matching and discrimination processing unit 123 determines that these voice features do not match (N), the controller 101 notifies the user to be authenticated that user authentication is not possible (S324).

On the other hand, if the voice feature matching and discrimination processing unit 123 determines that these voice features match (Y), the controller 101 determines that the user to be authenticated is a specific user (e.g., user A) and completes the user authentication process. The controller 101 then notifies the user to be authenticated that the user authentication process has been completed by displaying on the display unit 128 or emitting sound from the audio output processing unit 121 (S319). The user to be authenticated recognizes from the notification from the information processing device 100 that he or she has been authenticated as a registered user and that the user authentication process has been completed (S320). This ends the user authentication process.

<After user authentication process is completed>
The authenticated registered user (a particular user, for example, user A) is able to use various functions of the information processing device 100 (S340). Specifically, the authenticated registered user performs an input operation on the information processing device 100 (S341) to operate the wireless communication function, telephone function, and the like of the information processing device 100 (S342).

In addition, in step S341, when an authenticated registered user performs an input operation, such as tapping an icon, in step S342, the information processing device 100 executes the corresponding application to realize various functions.

The voice input processing unit 120 also captures voice (authenticated voice) uttered by a registered user who has been authenticated through a phone call, voice operation, etc., and converts it into a voice signal (fourth voice signal). The controller 101 then extracts the voice characteristics (fourth voice characteristics) of the authenticated registered user from this voice signal, and updates the voice characteristics of this registered user stored in the memory 114 based on the extracted voice characteristics.

Then, the voice password generation processing unit 123 generates a new voice password based on the updated voice characteristics and stores it in the memory 114. In this way, the information processing device 100 can generate more complex voice passwords by continually updating the voice characteristics for each registered user.

The information processing device 100 may also issue instructions to registered users who are currently using the device, via screen display or voice output, to have them speak more phrases. For example, the information processing device 100 may converse with registered users, or instruct registered users who use the device infrequently or who do not frequently perform voice input operations, to speak out loud. This enables the information processing device 100 to collect many voices from registered users, update the voice characteristics as needed, and generate more complex voice passwords. This improves the security of the information processing device 100.

<Scope of use of information processing device>
The information processing device 100 of the present embodiment is mounted on, for example, a head mounted display (HMD), and performs user registration processing and user authentication processing for a user of the HMD. The information processing device 100 may be, for example, a mobile phone, a smartphone, a tablet terminal, an AI speaker, etc.

<Effects, etc.>
According to this embodiment, in the user registration process, the controller 101 generates a voice password for the user to be registered based on the registered and stored voice characteristics, and in the user authentication process, the controller 101 has the user to be authenticated utter the voice password.

With this configuration, registered users do not need to associate or memorize their voice passwords, making it possible to perform user authentication using voice input in a user-friendly and highly secure manner.

Furthermore, according to this embodiment, the controller 101 generates multiple voice passwords for each registered user, and presents a voice password randomly selected from the multiple voice passwords to the user to be authenticated.

With this configuration, an audio password is automatically generated, eliminating the need for registered users to register their passwords, making it possible to further improve usability. In addition, with this configuration, the audio password presented to the user to be authenticated can be a one-time audio password, making it possible to prevent unauthorized authentication by third parties through eavesdropping or eavesdropping.

Furthermore, according to this embodiment, if the controller 101 determines that the presentation time that has elapsed since the audio password was presented to the user to be authenticated has exceeded the timeout time, the controller 101 stops the user authentication process.

It would take a considerable amount of time for a third party to obtain a one-time audio password by eavesdropping or eavesdropping. Therefore, this configuration makes it possible to further prevent unauthorized authentication by a third party using a one-time audio password.

In this embodiment, if the controller 101 determines that the voice password presented to the user to be authenticated matches the voice password spoken by the user to be authenticated, the controller 101 extracts a voice feature (third voice feature) from the voice password spoken by the user to be authenticated and determines whether this voice feature matches the voice feature of the specific user. However, the controller 101 is not limited to this sequence, and if the controller 101 determines that the voice password presented to the user to be authenticated matches the voice password spoken by the user to be authenticated, the controller 101 may determine that the user to be authenticated is a specific user and terminate the user authentication process.

This configuration reduces the time required for user authentication, and improves usability while ensuring the security of the user authentication process.

In this embodiment, voice features are extracted twice, when the authentication voice is captured (S303) and when the voice password is captured (S311), and compared with the registered and saved voice features (S314-S315). In this case, in step S303, the threshold for matching and discriminating the voice feature amount may be set broadly so as to narrow down the specific user to the already registered user with the closest feature amount, and in step S311, the threshold for matching and discriminating the voice feature amount may be set narrowly so that the user to be authenticated is determined to be the specific user only when the voice features match highly.

In addition, in this embodiment, the audio password generation processing unit 124 may change the complexity of the audio password as appropriate depending on the length of words or sentences, the number of audio passwords, etc. This configuration makes it possible to take into account the required security level and usage conditions, making it possible to ensure a good balance between safety and convenience.

In addition, in this embodiment, the voice password may be changed periodically, and a voice password that has been used once may not be reused. This makes it possible to further prevent unauthorized authentication by a third party, further improving safety.

The voice characteristics of a registered user may change depending on the time of day. For this reason, the information processing device 100 may register and store voice characteristics for each time period for each registered user. The controller 101 may then perform user authentication processing using a voice password that uses voice characteristics that match the time period. With this configuration, user authentication processing is performed using more appropriate voice characteristics, making it possible to improve the accuracy of user authentication.

In addition, since there is a possibility that the voice password of the user to be authenticated during the user authentication process may be intercepted by a third party outdoors, the voice password used outdoors may be used only indoors.

In addition, in this embodiment, a case has been described in which the audio password is automatically generated by the audio password generation processing unit 124, but audio generated by a registered user may also be registered as the audio password.

In addition, in this embodiment, a case has been described in which a randomly selected voice password is presented to the user, but multiple voice passwords may be presented and the user to be authenticated may select the voice password to be spoken from among them.

In addition, when the user authentication process starts, the user to be authenticated speaks some kind of voice for authentication, but a voice registered in advance by a registered user may also be spoken as the authentication voice.

(Embodiment 2)
Next, a second embodiment will be described. In this embodiment, a user authentication network system that performs user authentication between a server device and the system via a network will be described. In the following, descriptions of parts that overlap with the above-mentioned embodiment will be omitted as appropriate.

Figure 5 is a block diagram showing an example of the configuration of a user authentication network system according to embodiment 2 of the present invention. The user authentication network system 450 shown in Figure 5 includes information processing devices 400 (e.g., 400-1, 400-2), a server device 430, and an external network 440. The external network 440 connects the information processing devices 400 and the server device 430.

The information processing devices 400 (e.g., 400-1, 400-2) are included in the facilities 420 (e.g., 420-1 to 420-2). The facilities 420 (e.g., 420-1 to 420-2) each include, for example, an information processing device 400 (e.g., 400-1 to 400-2) and an entrance gate 410 (e.g., 410-1 to 410-2).

The entrance gate 410 is placed, for example, near the entrance of the facility 420. The information processing device 400 is also placed near the entrance gate 410 and performs user authentication processing on a person attempting to pass through the entrance gate 410 as an authentication target user. Once authenticated, the authentication target user is permitted to pass through the entrance gate 410 and can enter the facility 420.

The configuration of facility 420-2 is similar to that of facility 420-1, so the following describes the configuration of the facility using facility 420-1 as an example. Information processing device 400-1 is similar to the configuration of information processing device 100 in embodiment 1. The differences between information processing device 400-1 and information processing device 100 are that memory 114 does not include user information storage area 106, and that a functional operation control unit 401 has been added.

The functional operation control unit 401 is a functional block that controls the open/close state of the corresponding entrance gate 410-1. When the user authentication process is completed in the information processing device 400-1, the functional operation control unit 401 transmits an entrance gate control signal that switches the entrance gate 410 from a closed state to an open state. This allows the authenticated user to pass through the entrance gate 410-1.

The server device 430 is a device that manages information on registered users who are permitted to enter the facility 420. As shown in FIG. 4, the server device 430 includes a controller 421, a memory 424, a communication unit 429, and the like. Each unit that constitutes the server device 430 is connected to each other via a bus 428.

The memory 424 is a functional block that stores various information related to the server device 430. The memory 424 stores various information, such as various programs used by the controller 421, and voice characteristics and voice passwords for each registered user. The memory 424 includes a program storage area 423 and multiple user information storage areas 426. The user information storage area 426 corresponds to the user information storage area 106 in FIG. 1. Note that FIG. 4 shows user information storage areas 426e to 426g corresponding to users E to G, who are already registered users.

The program memory area 423 is a memory area that stores various programs such as the OS and operation control applications. These programs are deployed and executed by the controller 421.

The user information storage area 426e is a storage area that stores user information of user E. The user information storage area 426e includes, for example, a voice feature storage area 422e that stores the voice features of user E, and voice password storage areas 423e to 425e that store the voice password of user E. The voice password storage areas 423e to 425e each store a different voice password. Note that the number of voice password storage areas may be less than this, or more than this, as in the information processing device 100.

The user information storage area 426f is a storage area that stores user information of user F. The user information storage area 426f includes, for example, a voice feature storage area 422f that stores the voice features of user F, and voice password storage areas 423f to 425f that store the voice password of user F.

The user information storage area 426g is a storage area that stores user information for user G. Although not shown in the figure, the user information storage area 426g also includes a voice feature storage area that stores the voice features of user G, a plurality of voice password storage areas that store the voice passwords of user G, and the like.

The communication unit 429 is a communication interface that is connected to the external network 440 and transmits and receives information to and from the information processing device 400 via the external network 440. Specifically, the communication unit 129 of the information processing device 400 is connected to the communication unit 429 of the server device 430 via the external network 440, and information is transmitted and received between the communication units.

In addition, the information processing device 400 may transmit and receive information to and from the server device 430 via the base station 445. Specifically, the base station communication unit 130 of the information processing device 400 is connected to the communication unit 429 of the server device 430 via the base station 445 and the external network 440, and information is transmitted and received between the communication units.

The controller 421 is composed of a CPU and other components, and controls each component of the server device 430 by executing programs such as an OS and an operation control application stored in the memory 424, and performs operation control processing for the entire server device 430.

During user registration processing, the controller 421 receives the voice characteristics and voice password of the user to be registered sent from the information processing device 400 via the communication unit 429 and stores them in the memory 424. During user authentication processing, the controller 421 transmits the voice characteristics and voice password stored in the memory 424 to the information processing device 400 via the communication unit 429.

<User registration process>
Next, a description will be given of the user registration process in the user authentication network system 450. Note that, in the following, the user registration process will be described using the process between the information processing device 400-1 (facility 420-1) and the server device 430 as an example, but similar processes are also performed between the other information processing devices 400 and the server device 430.

Figure 6 is a diagram showing an example of an operational sequence of a user registration process according to the second embodiment of the present invention. In Figure 6, the operation of a user to be registered is shown on the left, the operation of the information processing device 400-1 is shown in the center, and the operation of the server device 430 is shown on the right. Note that in Figure 6, steps in which the same processes as those in Figure 2 are performed are given the same reference numerals, and detailed explanations of the corresponding steps will be omitted as appropriate.

In the user registration process of this embodiment, steps S201 to S207, S212 to S215, and the following steps S501 to S514 are performed. First, a description of each process from the input operation by the user to be registered to start the user registration process (S201) to the voice feature extraction process by the information processing device 400-1 (S207) will be omitted.

When the extraction process of the voice feature (first voice feature) is performed, the information processing device 400-1 transmits a voice feature transmission request to the server device 403 to transmit the voice features of all registered users stored in the server device 430 to its own information processing device 400-1 (S501).

When the server device 403 receives a request to send the voice characteristics of registered users sent from the information processing device 400-1 (S502), it sends the voice characteristics of all registered users stored in the memory 424 to the information processing device 400-1 (S503). Note that if the server device 430 manages registered users for each information processing device 400 (facility 420), it will only send the voice characteristics of registered users of the information processing device (e.g., 400-1) that made the request to send the voice characteristics.

When the information processing device 400-1 receives the voice features sent from the server device 403 (S504), the voice feature matching and discrimination processing unit 123 compares the voice features extracted in step S207 with the received voice features of the registered user (S505) and determines whether the extracted voice features have already been registered (S506). If it is determined that the extracted voice features do not match the voice features of any registered user (N), the controller 101 determines that the user to be registered is a new user who has not been registered, and transmits the extracted voice features to the server device 430 (S507).

When the server device 430 receives the voice features transmitted from the information processing device 400-1, it registers the received voice features in the user information storage area 426 in the memory 424 where no user information is registered (S508). Specifically, the controller 421 assigns a new user information storage area 426 for this user to be registered, and stores the voice features of the user to be registered in the voice feature storage area of the newly assigned user information storage area 426. The server device 430 then transmits a storage completion notification to the information processing device 400-1 notifying that the registration and storage of the received voice features has been completed (S509).

When the information processing device 400-1 receives a notification that registration of the transmitted voice characteristics has been completed (S510), it generates multiple corresponding voice passwords based on the registered voice characteristics and transmits them to the server device 403 (S511).

The voice password generation processing unit 124 generates multiple voice passwords for the newly registered user based on the registered voice characteristics. The information processing device 400-1 then transmits the multiple generated voice passwords to the server device 430 (S511).

When the server device 430 receives the multiple voice passwords sent from the information processing device 400-1, it stores the received voice passwords in the voice password storage area of the user information storage area 426 corresponding to the newly registered user (S512), and sends a registration completion notification to the information processing device 400-1 indicating that the voice passwords have been registered and stored (S513).

When the information processing device 400-1 receives the notification that the voice password registration has been completed (S514), the user registration process is completed. The controller 101 of the information processing device 400-1 notifies the user to be registered (newly registered user) that the user registration has been completed by displaying the message on the display unit 128 and emitting sound from the voice output processing unit 121 (step S212).

On the other hand, in step S506, if the extracted voice characteristics of the user to be registered match any of the voice characteristics transmitted from the server device 430 (Y), the controller 101 determines that the user to be registered is an already registered user, and notifies the user to be registered that he or she has already been registered by displaying a message on the display unit 128 or emitting a sound from the audio output processing unit 121 (step S214). Note that a detailed description of steps S212 to S215 will be omitted.

<User authentication process>
Next, the user authentication process in this embodiment will be described. Note that, although the user authentication process will be described below using the process between the division processing device 400-1 (facility 420-1) and the server device 430 as an example, the same process is also performed between other information processing devices 400 (facility 420) and the server device 430.

Figure 7 is a diagram showing an example of an operation sequence of a user authentication process according to the second embodiment of the present invention. In Figure 7, the operation of the user to be authenticated is shown on the left, the operation of the information processing device 400-1 is shown in the center, and the operation of the server device 430 is shown on the right. Note that in Figure 7, steps in which the same processes as those in Figures 3 to 4 are performed are given the same reference numerals, and detailed explanations of the corresponding steps will be omitted as appropriate. Note that the main sequence S350 in Figure 7 is the same as in Figure 4, and only steps showing the connection relationship between the main sequence and outside the main sequence are shown for convenience.

First, we will omit explanations of steps S301 to S303, which start when the user authentication process begins, from when the user to be authenticated speaks an authentication voice (S301) to when the information processing device 400 extracts the voice characteristics of the user to be authenticated from the voice signal of the authentication voice (S303).

Then, when the voice characteristics of the user to be authenticated are extracted, the information processing device 400-1 transmits a voice characteristic transmission request to the server device 430 to transmit the voice characteristics registered in the server device 430 to its own information processing device 400-1 (S601).

When the server device 430 receives a request to transmit voice characteristics from the information processing device 400 (S602), it transmits the voice characteristics registered in the memory 424 to the information processing device 400-1 (S603).

When the information processing device 400-1 receives the voice characteristics transmitted from the server device 403, it temporarily stores the received voice characteristics, for example, in the RAM or memory 114 in the controller 101 (S604).

The voice feature matching and discrimination processing unit 123 compares the voice features extracted in step S303 with the voice features of already registered users received from the server device 430 (S304), and determines whether the extracted voice features have already been registered (S305).

If the voice feature matching and discrimination processing unit 123 determines that the extracted voice feature of the user to be authenticated matches any of the voice features received from the server device 430 (Y), the controller 101 narrows down the registered users (e.g., user E) corresponding to the voice features that match those of the user to be authenticated as specific users to be authenticated by voice password (S306). Once the specific users (e.g., user E) have been narrowed down, the information processing device 400-1 transmits a request to send the specific user's voice password to the server device 430 (S605).

When the server device 430 receives a request to send a voice password from the information processing device 400-1 (S606), the controller 421 reads all voice passwords from the multiple voice password storage areas (423e to 425e) corresponding to user E, and transmits the read voice passwords to the information processing device 400-1 via the communication unit 429 (S607).

When the information processing device 400-1 receives the voice password transmitted from the server device 430, it temporarily stores the received voice password, for example, in the RAM or memory 114 in the controller 101 (S608).

Then, the controller 101 randomly selects one voice password from the received voice passwords of the specific user (user E) (S307).

On the other hand, in step S305, if the voice feature matching and discrimination processing unit 123 determines that the extracted voice feature does not match any of the voice features received from the server device 430 (N), the controller 101 notifies the user to be authenticated that user authentication is impossible by displaying a message on the display unit 128 or emitting a sound from the voice output processing unit 121 (S324).

From steps S307 and S324 onwards, each process included in the main sequence S350 in FIG. 4 is performed. Among these, in step S320, the user to be authenticated may notify the information processing device 400-1 that he/she has recognized the completion of the user authentication process, for example, by touching a confirmation button displayed on the display unit 128.

When the user to be authenticated recognizes that the user authentication process has been completed, the function operation control unit 401 transmits an entrance gate control signal that switches the entrance gate 410-1 from a closed state to an open state. This opens the entrance gate 410-1, and the authenticated user to be authenticated can pass through the entrance gate 410-1 and enter the facility 420-1 (S610).

Note that step S610 may be performed after a predetermined waiting time has elapsed since the completion of user authentication was notified in step S319.

<Effects, etc.>
According to this embodiment, in addition to the effects of the first embodiment already described, the following effects can be obtained. According to this embodiment, the voice characteristics and voice passwords of registered users are stored in a server device 430 provided at a location remote from the information processing device 400, and the information processing device 400 receives the voice characteristics and voice passwords from the server device 430 via the external network 440. According to this configuration, even from the information processing device 400 in any facility 420 connected to the external network 440, it is possible to receive the voice characteristics and voice passwords stored in the server device 430 and perform user authentication processing. Furthermore, management of various information of registered users is centralized.

In addition, in this embodiment, a cloud server may be used as the server device. With this configuration, the capacity of the server device 430 is increased, making it possible to store information on a larger number of already registered users. In addition, it becomes possible to build and utilize a database of already registered users.

Also in this embodiment, the information processing device 400 may capture the voice uttered by the registered user after the user authentication process, update the voice characteristics as needed, and generate a new voice password based on the updated voice characteristics. The updated voice characteristics and the new voice password are transmitted to the server device 430 and stored in the memory 424.

The information processing device 400 may store the generated audio password in its own memory 114, taking into consideration the capacity of the server device 430, the communication status with the server device 430, etc.

In addition, in this embodiment, the opening control of the entrance gate 410 is exemplified as the user authentication network system, but the user authentication network system is not limited to this and can also be used for places where login processing is performed using a personal computer or the like, or where entry management for multiple rooms is performed, etc.

In the examples shown in Figures 1 and 4, the voice spoken by the user to be registered or the user to be authenticated is used at the start of the user registration process or the user authentication process, but data information such as a user ID (identification) that identifies the user may be input instead of this voice.

In this case, during user registration processing, the information processing device 400 checks the input data information to determine whether the user to be registered is a previously registered user, and if it determines that the user to be registered is not a previously registered user, notifies the user to utter some kind of voice (which may be a specified voice password). The information processing device 400 captures the voice uttered by the user, extracts voice characteristics from this voice, generates a voice password, and registers the user to be registered together with the voice characteristics and voice password.

In addition, during user authentication processing, the information processing device 400 checks the input data information to determine whether the user to be authenticated is a registered user, and if it determines that the user is a registered user, it randomly selects an audio password of the registered user and presents it to the user to be authenticated.

The voice input processing unit 120 shown in Figs. 1 and 4 is configured to capture spoken voice using a microphone, but there are other methods for capturing the user's voice, such as bone conduction. Bone conduction is sound that travels through the air, vibrates the eardrum, and is transmitted to the auditory nerve; in other words, bone conduction sound is sound that travels through the skull and is transmitted directly to the auditory nerve by vibrations of the vocal cords, as opposed to sound collected from the ear. For example, a head-mounted display that is worn on the head can capture not only air conduction sound, but also bone conduction sound through bone conduction, as the display is worn on the head. In this case, user registration and authentication are performed in the same manner as in the embodiment shown in Figs. 1 and 4.

The present invention is not limited to the above-described embodiment, but includes various modifications. The above-described embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to having all of the configurations described.

It is also possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace part of the configuration of each embodiment with other configurations. Note that the components and relative sizes shown in the drawings have been simplified and idealized to explain the invention in an easy-to-understand manner, and the actual shape may be more complex.

100, 400... Information processing device, 101... Controller, 114... Memory, 120... Voice input processing unit, 424... Memory, 430... Server device, 440... External network, 450... User authentication network system

Claims (4)

1. A method for authenticating a user , comprising:
a storage step of storing a first voice feature extracted from a voice uttered by a user and a first voice password generated based on the first voice feature in association with the user;
an authentication processing step of performing a user authentication process for the user using the first voice password stored in the storage step;
a voice capture step of capturing a post-authentication voice uttered by the user after the user authentication process is completed by the authentication process step ;
a generating step of extracting a second voice feature from the post-authentication voice captured in the voice capturing step , and generating a second voice password based on the extracted second voice feature;
a management step of updating the first voice characteristic and the first voice password stored in the storage step to the second voice characteristic and the second voice password .
User authentication method. 2. The user authentication method according to claim 1,
The storing step includes storing, as the first voice password, a voice password randomly selected from a plurality of voice passwords generated based on the first voice characteristic.
User authentication method. 3. The user authentication method according to claim 1,
the authentication processing step presents the first voice password stored in the storage step to the user, prompts the user to speak the first voice password, and performs a user authentication process for the user by comparing and matching an extracted password extracted from the user's speech with the first voice password;
User authentication method. 4. The user authentication method according to claim 3 ,
the authentication processing step stops the user authentication processing when a voice uttered by the user is not captured even after a timeout time has elapsed since the first voice password was presented to the user;
User authentication method.

Images