JP2023155684A - Program, information processing device and method - Google Patents

Abstract

To achieve voice authentication with high safety at high speed.SOLUTION: A program to be executed by a computer including a processor and a memory, causes the processor to execute the steps of: extracting a first voice feature amount from first voice data of a pre-registered user; generating a temporary password; presenting the password to the user; receiving an input of second voice data of the user reading out the password; extracting a second voice feature amount from the received second voice data; and executing user authentication using the first voice feature amount, the second voice feature amount and the password, where the first voice feature amount and the second voice feature amount are voice features represented by vectors.SELECTED DRAWING: Figure 7

Description

The present disclosure relates to a program, an information processing device, and a method.

In order to achieve safe and reliable dual identity authentication, we learn and register a user voiceprint model in advance, generate a dynamic password, and create a global There is a technique that calculates the overall identity reliability of the requester using a character acoustic model and a user voiceprint model, and determines the identity of the requester based on the calculated overall identity reliability (Patent Document 1).

Special table 2018-509649 publication

However, with conventional technology, it is necessary to learn a global voiceprint model, use a calculation method used for conventional speech recognition such as the Viterbi algorithm for reliability, or calculate an average reliability value, which makes processing difficult. The problem was that it was slow.

An object of the present disclosure is to quickly realize highly secure voice authentication.

A program according to an aspect of the present disclosure is a program to be executed by a computer including a processor and a memory, and the program causes the processor to select first voice data of a user registered in advance. a step of extracting a voice feature amount, a step of generating a temporary password, a step of presenting the password to the user, a step of receiving input of second voice data in which the user reads out the password; extracting a second audio feature from the received second audio data; and performing user authentication using the first audio feature, the second audio feature, and the password. The first audio feature amount and the second audio feature amount are audio features represented by vectors.

According to the present disclosure, highly secure voice authentication can be realized at high speed.

FIG. 1 is a block diagram showing the configuration of an information processing system 1 of the present disclosure. FIG. 1 is a block diagram showing the configuration of an information processing device 10 of the present disclosure. 1 is a block diagram showing a functional configuration of an information processing device 10 of the present disclosure. FIG. It is a flowchart which shows the 1st voice data collection process of this indication. It is a flowchart which shows learning processing of this indication. It is a flowchart which shows the 1st voice feature quantity extraction process of this indication. 2 is a flowchart showing authentication processing according to the present disclosure. It is a flowchart which shows user authentication processing of authentication processing S405 of this indication. FIG. 1 is a block diagram showing the configuration of an information processing system 2 of the present disclosure. 2 is a flowchart showing authentication processing according to the present disclosure. FIG. 1 is a block diagram showing the configuration of an information processing system 3 of the present disclosure. 2 is a flowchart showing authentication processing according to the present disclosure. 1 is a block diagram showing the configuration of an information processing system 4 of the present disclosure. FIG. 2 is a flowchart showing authentication processing according to the present disclosure. 1 is a block diagram showing the configuration of an information processing system 5 of the present disclosure. FIG. 2 is a flowchart showing authentication processing according to the present disclosure. 1 is a block diagram showing the configuration of an information processing system 6 of the present disclosure. FIG. 2 is a flowchart showing authentication processing according to the present disclosure.

Hereinafter, one embodiment of the present disclosure will be described in detail based on the drawings. In addition, in the drawings for explaining the embodiments, the same components are generally designated by the same reference numerals, and repeated explanations thereof will be omitted.

Conventional technology requires learning a global voiceprint model, uses a calculation method used for conventional speech recognition such as the Viterbi algorithm for reliability, and calculates an average reliability value, which slows down processing speed. The problem was that it was slow.

The technology of the present disclosure extracts a first voice feature amount from first voice data of a user registered in advance, generates a temporary password, and presents the password to the user. Then, a second audio feature is extracted from the second audio data in which the user reads out the password, and the user is authenticated using the first audio feature, the second audio feature, and the password. I do. Further, the first audio feature amount and the second audio feature amount are audio features expressed by vectors. Accordingly, the present disclosure discloses a technology that can quickly realize highly secure voice authentication.

Further, the conventional technology has problems in that the processing speed is slow and it is difficult for service providers, service users, etc. to use the technology. For example, in accommodation services such as hotels, it may be necessary to check in manually at the front desk, or even if the user does not want to be aware of the authentication process, the user may notice that the user has authenticated. There was a problem. The present disclosure discloses a voice authentication technology that is highly convenient depending on the usage scene.
In the first embodiment, the voice authentication technology of the present disclosure will be described below. Furthermore, in the second to sixth embodiments, specific examples of highly convenient voice authentication techniques according to usage scenes will be described.

<First embodiment>
(1) Configuration of information processing system 1 FIG. 1 is a block diagram showing the configuration of information processing system 1 according to the first embodiment. As shown in FIG. 1, the information processing system 1 includes an information processing device 10, a user terminal 20, and a network 30. The information processing device 10 and the user terminal 20 are connected to be able to communicate with each other via a network 30 using a wired or wireless communication standard.

The information processing device 10 is realized by a stationary PC (Personal Computer), a laptop PC, or the like.

FIG. 2 is a block diagram showing the configuration of the information processing device 10 of the first embodiment. As shown in FIG. 2, the information processing device 10 includes a storage device 11, a processor 12, an input/output interface 13, and a communication interface 14.

The storage device 11 is a device that temporarily stores programs, data processed by the programs, and the like. The storage device 11 is realized by one or a combination of a flash memory, a memory such as a DRAM (Dynamic Random Access Memory), an HDD (Hard Disc Drive), an SSD (Solid State Drive), and the like.

The processor 12 is hardware for executing a set of instructions written in a program, and is composed of an arithmetic unit, registers, peripheral circuits, and the like.

The input/output interface 13 is an interface that receives input signals from an input device (for example, a microphone, a touch panel, a touch pad, a pointing device such as a mouse, a keyboard, etc.) that is not shown. Further, the input/output interface 13 is an interface that transmits an output signal to an output device (not shown) (display, speaker, etc.).

The communication interface 14 is an interface for inputting and outputting signals so that the information processing device 10 communicates with an external device.

The user terminal 20 is a terminal device operated by or for the user. The user is, for example, a user of a service. When operated by a user, the user terminal 20 is a terminal device owned by the user, or a terminal device provided to the user by a service provider and owned by the provider. When the user terminal 20 is a terminal device operated for a user, the user terminal 20 is a terminal device owned by a service provider.

The user terminal 20 is realized by, for example, a smartphone compatible with a mobile communication system, a mobile terminal such as a tablet, a wearable device, or the like. In addition, the user terminal 20 may be a stationary PC (Personal Computer), a laptop PC, or the like. In this disclosure, a case where the user terminal 20 is a smartphone will be described as an example.

FIG. 2 is a block diagram showing the configuration of the user terminal 20. As shown in FIG. 2, the user terminal 20 includes a storage device 21, a processor 22, an input/output interface 23, and a communication interface 24. Further, the user terminal 20 includes an output device such as a display and a speaker (not shown).

The storage device 21 is a device that temporarily stores programs and data processed by the programs. The storage device 11 is realized by one or a combination of a flash memory, a memory such as a DRAM (Dynamic Random Access Memory), an HDD (Hard Disc Drive), an SSD (Solid State Drive), and the like.

The processor 22 is hardware for executing a set of instructions written in a program, and is composed of an arithmetic unit, registers, peripheral circuits, and the like.

The input/output interface 23 is an interface that receives input signals from an input device (for example, a microphone, a touch panel, a touch pad, a pointing device such as a mouse, a keyboard, etc.) that is not shown. Further, the input/output interface 13 is an interface that transmits an output signal to an output device (not shown) (display, speaker, etc.).

The communication interface 24 is an interface for inputting and outputting signals so that the information processing device 10 communicates with an external device.

(2) Functions of the information processing device 10 FIG. 3 is a block diagram showing the functional configuration of the information processing device 10 of the first embodiment. As shown in FIG. 3, the information processing device 10 includes a communication section 110, a storage section 120, and a control section 130.

The communication unit 110 performs processing for the information processing device 10 to communicate with an external device.

The storage unit 120 stores data and programs used by the information processing device 10. The storage unit 120 stores a first data DB 121, a second data DB 122, a third data DB 123, and the like.

The first data DB 121 is a database for holding first audio data. For example, the first data DB 121 includes records such as item "ID", item "user ID", and item "first audio data". Note that the items shown here are not all, and there may be other items.

The item "ID" stores information for identifying each record.

The item "user ID" stores information for identifying a user. Note that the user ID is the same in other DBs of the present disclosure.

The item "first audio data" stores audio data generated by the user. The audio data is, for example, data represented by an audio file such as WAV.

The second data DB 122 is a database that holds learned models and parameters of the learned models. The trained model will be described later.

The third data DB 123 is a database for holding the user's first voice feature amount, which will be described later. For example, the third data DB 123 includes records such as the item "user ID", the item "user name", the item "audio feature amount", and the item "update date and time". Note that the items shown here are not all, and there may be other items.

The item "user name" stores information such as the user's first name, last name, full name, name, and nickname.

The item "voice feature amount" stores the extracted first voice feature amount of the user. The first audio feature amount will be described later.

The item "update date and time" holds the date and time when the first audio feature amount was stored in the third data DB 123.

The control unit 130 performs processing according to a program by the processor 12 of the information processing device 10 to provide information to the reception control unit 131, transmission control unit 132, extraction unit 133, generation unit 134, presentation unit 135, authentication unit 136, etc. Demonstrate function.

The reception control unit 131 controls a process in which the information processing device 10 receives a signal from an external device according to a communication protocol. For example, upon receiving second voice data in which a password, which will be described later, is read out from the user terminal 20, the reception control unit 131 passes the second voice data to the extraction unit 133.

The transmission control unit 132 controls a process in which the information processing device 10 transmits a signal to an external device according to a communication protocol.

The extraction unit 133 extracts voice features from the voice data.
Specifically, the extraction unit 133 extracts the voice feature amount using the learned model and one or more first voice data of the user registered in the first data DB 121.

Here, the learned model of the present disclosure will be explained. The learned model is a model that has been trained to output a voice feature amount representing a user's voice feature as a vector in response to input voice data. As the trained model, any machine learning model, any neural network, etc. can be used.

In this disclosure, a case will be described in which the learned model is a deep metric learning model. In this case, the trained model is trained to output a voice feature represented by an N-dimensional vector when input with a voice feature obtained by converting voice data into a two-dimensional feature. The audio feature amount is, for example, a mel frequency cepstral coefficient (MFCC) of the audio. The learned model may be trained by the information processing device 10 or by another device. In the present disclosure, an example will be described in which a trained model is trained by another device and stored in the second data DB 122 in advance.

The learning data used for learning the trained model is prepared in advance by performing at least one of audio signal processing and acoustic processing on voice data of a plurality of people recorded for learning. The audio signal processing and acoustic processing include, for example, volume adjustment, audio expansion/contraction, pitch shift, noise application, equalizer, reverb, and the like. A plurality of pieces of audio data can be generated from a person's voice data by subjecting the person's voice data to a plurality of audio signal processes and acoustic processing. Such audio signal processing and audio processing are performed to take into account differences in the recording environment of audio data. With such training data, the trained model can realize the extraction of audio features with less influence of the recording environment such as differences in microphone performance.

Further, the processed learning data is converted into a feature expressed by a two-dimensional vector using Mel frequency cepstral coefficients (MFCC). The feature amount is expressed, for example, in two dimensions of 128×128.

The trained model is then trained to input the feature amount represented by the two-dimensional vector and output the audio feature amount represented by the N-dimensional vector. N is any integer. The voice feature quantity represented by an N-dimensional vector is a vector whose elements are what kind of voice characteristics a person has. The learning method uses distance learning, for example. As a result, the trained model is trained to output a voice feature quantity represented by an N-dimensional vector so as to indicate the voice characteristics of each person. Since the audio feature amount output in this way is a vector representation, it is possible to accurately determine whether or not they are the same person based on the distance from other similarly output audio feature amounts. The learned model learned in this way is stored in the second data DB 122.

Note that the learning data may include adversarial noise. The learned model can be made resistant to adversarial attacks (such as spoofing) by performing adversarial learning during learning of the trained model.

Specifically, the extraction unit 133 acquires the user's first voice data from the first data DB 121. Next, the extraction unit 133 converts the first audio data into a third audio feature represented by a two-dimensional vector using Mel frequency cepstral coefficients (MFCC). Next, the extraction unit 133 acquires the learned model from the second data DB 122. Next, the extraction unit 133 uses the third audio feature and the learned model to extract the first audio feature represented by an N-dimensional vector. Then, the extraction unit 133 stores the extracted first voice feature amount in the third data DB 123.

Further, when the extraction unit 133 receives second voice data that is the voice of the user reading out the password described below, the extraction unit 133 converts the second voice data into a two-dimensional vector using Mel frequency cepstral coefficients. It is converted into a fourth audio feature amount. Next, the extraction unit 133 uses the fourth voice feature and the learned model to extract a second voice feature expressed by an N-dimensional vector.

The generation unit 134 generates a temporary password.

Specifically, the generation unit 134 generates a password with a predetermined expiration date. The password is a character string that can be read aloud. The generating unit 134 may generate the password as a random character string, or may generate the password as a password selected from predetermined words or sentences at random or in a predetermined manner. If the password is a random string of characters, it may be difficult for a person to read it out loud. Therefore, if the generation unit 134 selects a password in advance from words or sentences that are easy to read out, the accuracy of user authentication using voice becomes higher.

The presentation unit 135 presents the password to the user.

Specifically, the presenting unit 135 presents the password to the user in a manner that is perceivable by a person and can be uttered out loud as a result of the percept. For example, the presentation unit 135 causes the output device to display the password as a character string, displays the password as a recognizable image or video, or causes the output device to produce a sound related to the password. In the case of displaying the image or video, the presentation unit 135 displays the image or video of the Big Dipper on an output device (for example, a display) if the password is "Hokutoshichisei", for example. In addition, in the case where the password is pronounced as a sound, for example, if the password is "Hokutoshichisei", the presentation unit 135 generates the sound read out by an arbitrary method and causes the output device (for example, a speaker) to pronounce the sound. .

Furthermore, if the user is not near the output device connected to the information processing device 10, the presenting unit 135 may present the password to the user by transmitting the password to the user terminal 20 or the like via communication. .

The authentication unit 136 performs user authentication using the first audio feature amount, the second audio feature amount, and the password.

Specifically, the authentication unit 136 performs password authentication and speaker authentication when the password is within the expiration date, and if both authentications are successful, it is determined that the user authentication has been successful. Password authentication is to authenticate that the second audio data is audio data that is read aloud regarding the password. Speaker authentication is to authenticate that the second voice data is uttered by a registered user.

First, the authentication unit 136 converts the received second audio data into text data. Next, the authentication unit 136 performs password authentication using the password and text data. More specifically, the authentication unit 136 verifies whether the text data obtained by converting the second audio data and the password match. If the text data and the password match, the authentication unit 136 determines that the password authentication is successful. On the other hand, if the text data and the password do not match, the authentication unit 136 determines that the password authentication has not been successful.

Further, the authentication unit 136 performs speaker authentication according to the distance between the first voice feature amount and the second voice feature amount. More specifically, the authentication unit 136 first calculates the distance between the first audio feature amount and the second audio feature amount. Next, the authentication unit 136 determines whether the calculated distance is less than or equal to a predetermined threshold. If the distance is less than or equal to a predetermined threshold, the authentication unit 136 determines that speaker authentication has been successful. On the other hand, if the distance is not less than a predetermined threshold, the authentication unit 136 determines that speaker authentication has not been successful.

The authentication unit 136 determines that user authentication has been successful if the password is within the expiration date and both password authentication and speaker authentication are successful. On the other hand, if the password is not within the expiration date, password authentication is not successful, or speaker authentication is not successful, the authentication unit 136 determines that user authentication has not been successful.

Note that the authentication unit 136 does not need to use the distance itself between the first audio feature amount and the second audio feature amount. For example, the authentication unit 136 may perform user authentication using the user's anomaly detection model learned using the first voice feature and the second voice feature. The user's anomaly detection model is a model that receives a voice feature as input and outputs whether or not the voice feature is the voice feature of the user. As the abnormality detection model, for example, existing models such as OneClassSVM and IsolationForest can be used. Note that the anomaly detection model is learned for each user using voice information at the time of user registration.

The authentication unit 136 then outputs the authentication result. The output destination of the authentication result is, for example, an output device connected to the information processing device 10, a server for transmitting the authentication result to the user terminal 20, and implementing a service.

(3) Operation In the following, processing in the information processing device 10 will be described with reference to the drawings.

FIG. 4 is a flowchart illustrating an example of the flow of the first audio data collection process performed by the information processing device 10. The information processing device 10 executes this process at an arbitrary timing (for example, at the timing when the first audio data is received).

In step S101, the reception control unit 131 receives first audio data from the user terminal 20.

In step S102, the reception control unit 131 stores the received first audio data in the first data DB 121, and ends the process.

FIG. 5 is a flowchart showing an example of the flow of learning processing performed by the information processing device 10. As shown in FIG. The process is executed at an arbitrary timing (for example, at the timing when a signal for starting the learning process is received). Note that when the information processing device 10 performs learning processing, a learning unit (not shown) performs the processing.

In step S201, the learning unit acquires learning data. In this case, the learning section acquires the learning data from the storage section 120 or an external database.

In step S202, the learning unit learns the model so as to output a voice feature amount representing the user's voice feature as a vector in response to input of voice data.

In step S203, the learning unit stores the learned model and its parameters in the second data DB 122, and ends the process.

FIG. 6 is a flowchart illustrating an example of the flow of the first audio feature amount extraction process performed by the information processing device 10. The information processing device 10 executes this process at an arbitrary timing (for example, the timing when the user's first voice data is stored, the timing when the authentication request signal is received, etc.).

In step S301, the extraction unit 133 acquires the user's first voice data from the first data DB 121.

In step S302, the extraction unit 133 converts the first audio data into a third audio feature represented by a two-dimensional vector using Mel frequency cepstral coefficients (MFCC).

In step S303, the extraction unit 133 acquires the learned model from the second data DB 122.

In step S304, the extraction unit 133 uses the third audio feature and the trained model to extract the first audio feature expressed by an N-dimensional vector.

In step S305, the extraction unit 133 stores the extracted first voice feature amount in the third data DB 123, and ends the process.

FIG. 7 is a flowchart illustrating an example of the flow of user authentication processing performed by the information processing apparatus 10. The information processing device 10 executes this process at an arbitrary timing (for example, at the timing when the authentication request signal is received).

In step S401, the reception control unit 131 receives an authentication request signal. The authentication request signal receives input from, for example, an input terminal connected to the information processing device 10, the user terminal 20, or the like.

In step S402, the generation unit 134 generates a temporary password.

In step S403, the presenting unit 135 presents the password to the user.

In step S404, the reception control unit 131 receives the second audio data.

In step S405, the authentication unit 136 performs user authentication using the first audio feature, the second audio feature, and the password.

In step S406, the authentication unit 136 outputs the authentication result and ends the process.

FIG. 8 is a flowchart illustrating an example of the flow of authentication processing performed by the authentication unit 136 in step S405.

In step S451, the extraction unit 133 converts the second audio data into a fourth audio feature represented by a two-dimensional vector using Mel frequency cepstral coefficients.

In step S452, the extraction unit 133 uses the fourth voice feature and the trained model to extract a second voice feature expressed by an N-dimensional vector.

In step S463, the authentication unit 136 acquires the first voice feature amount of the user related to the authentication request signal from the third data DB 123.

In step S464, the authentication unit 136 converts the received second audio data into text data.

In step S465, the authentication unit 136 performs password authentication using the password and text data.

In step S466, the authentication unit 136 performs speaker authentication according to the distance between the first voice feature and the second voice feature.

In step S467, the authentication unit 136 determines that user authentication is successful if the password is within the expiration date and both password authentication and speaker authentication are successful; otherwise, user authentication is successful. It is assumed that this has not been done, and the authentication result is returned.

Note that although the above processing has been described as individual processing, it is not limited to this. For example, the information processing system 1 may execute a combination of the above processes.

(4) Summary As explained above, according to the present disclosure, the first voice feature is extracted from the user's first voice data registered in advance, a temporary password is generated, and the user is given the following information: Present the password. Then, a second audio feature is extracted from the second audio data in which the user reads out the password, and the user is authenticated using the first audio feature, the second audio feature, and the password. I do. Further, the first audio feature amount and the second audio feature amount are audio features expressed by vectors. Accordingly, the present disclosure discloses a technology that can quickly realize highly secure voice authentication.

The trained model is a convolutional neural network such as a deep distance learning model, and the first audio feature amount and the second audio feature amount are represented by a predetermined N-dimensional vector. As a result, it is possible to employ a simple calculation of calculating the distance between the first voice feature amount and the second voice feature amount, so that highly secure voice authentication can be realized at high speed.

Note that although the password has been described using an example where the password is generated by the generation unit, the password is not limited to this. A configuration may also be adopted in which the password is acquired from the terminal that generated it. The same applies to other embodiments.

<Second embodiment>
In the second embodiment, an example will be described in which the above user authentication is used for authentication when using a service. In the second embodiment, a case will be described in which a service provides facilities of a service provider to a user. Note that the same configurations as those in the first embodiment are given the same reference numerals and the description thereof will be omitted.

(1) Configuration of the information processing system 2 FIG. 9 is a block diagram showing the configuration of the information processing system 2 of the second embodiment.

As shown in FIG. 9, the information processing system 2 includes an information processing device 10, a user terminal 20, a network 30, and a facility 40. The information processing device 10, the user terminal 20, and the facility 40 are connected to be able to communicate with each other via the network 30 using a wired or wireless communication standard.

Facility 40 is a facility provided to users by a service provider. The facility 40 is, for example, a sports gym, a pool, a bathing facility, an office, an accommodation facility, or the like. The facility 40 includes an information processing device 41 and a voice input device 42 installed in a predetermined facility. The information processing device 41 and the voice input device 42 are connected to be able to communicate with each other using a wired or wireless communication standard. For example, if the facility 40 is a sports gym, the predetermined equipment is a training room, a training machine, etc. Further, the information processing device 41 is communicably connected to the information processing device 10 and the like via the network 30 using a wired or wireless communication standard.

The information processing device 41 is, for example, an information processing device with a display. The information processing device 41 has the following functions.
- A function that detects whether there is a person in front of the information processing device 41 using infrared rays or the like.
- A function of acquiring a password by communicating with the information processing device 10 when it is detected that there is a person in front of the information processing device 41.
- A function to output the password to an output device (for example, a display) connected to the information processing device 41.
- A function of transmitting the second audio data acquired from the audio input device 42 to the information processing device 10.
- A function of unlocking a predetermined facility or activating a device related to providing a service in response to receiving permission information, which will be described later, from the information processing device 10.

The audio input device 42 is a device for the user to input second audio data. For example, the audio input device 42 has a microphone and converts audio input into the microphone into second audio data. The audio input device 42 passes the second audio data to the information processing device 41.

(2) Functions of the information processing device 10

In this embodiment, the first audio data stored in the first data DB 121 is acquired before the user starts using the service. For example, if the service is a gym, the first audio data is acquired at the timing when the user and the service provider conclude a membership contract.

The presentation unit 135 presents a password to the user when the user uses the service. Specifically, when the reception control unit 131 receives a password request from the information processing device 41, the presentation unit 135 transmits the password to the information processing device 41. Thereby, the presenting unit 135 presents the password to the user using an information processing device installed in a predetermined equipment within the facility 40.

When the reception control unit 131 receives the second audio data from the information processing device 41, the authentication unit 136 performs user authentication using the first audio feature amount, the second audio feature amount, and the password. . The authentication unit 136 permits use of the service in response to successful user authentication. Specifically, if the authentication unit 136 succeeds in user authentication, the authentication unit 136 transmits permission information for unlocking a predetermined facility or activating a device related to providing a service to the information processing device.

(3) Operation In the following, processing in the information processing device 10 will be described with reference to the drawings.

FIG. 10 is a flowchart illustrating an example of the flow of authentication processing performed by the information processing apparatus 10. The information processing device 10 executes this process at the timing when the password request is received from the information processing device 41, etc.

In step S501, the reception control unit 131 receives a password request from the information processing device 41.

In step S503, the presenting unit 135 presents a password to the user when the user uses the service.

In step S506, if the user authentication is successful, the authentication unit 136 transmits permission information for unlocking a predetermined facility or activating a device related to providing a service to the information processing device 41, and ends the process. do.

(4) Summary According to the present disclosure, use of the service is permitted in response to successful user authentication, and the first audio data is obtained before the user starts using the service. , when the user uses the service, the password is presented to the user. Thereby, highly convenient voice authentication can be realized depending on the usage scene. For example, in the present disclosure, user authentication can be performed using only voice. For this reason, if the user is busy with baggage, does not want to carry a physical key, does not want to check it in, or does not want to give it to the user, does not want to send a physical key, or does not want to make multiple physical keys, etc. can solve the problem. Furthermore, the present disclosure is highly convenient for the user because the user does not have to remember a password.

Further, the password is presented to the user by the information processing device installed in a predetermined equipment within the facility, and the input of the second audio data is accepted through the audio input device installed in the equipment. Then, in response to successful user authentication, the device unlocks the facility or activates the device related to providing the service. This allows the user to enter the facility and the device to be activated and available for use without the need for the service provider to deploy reception staff or the like. Therefore, the burden on the service provider can also be reduced.

<Third embodiment>
In the third embodiment, an example will be described in which the above user authentication is used for authentication when using a service. In the third embodiment, a case will be described in which the service is provision of use of accommodation facilities. Note that the same configurations as those in the first embodiment and the second embodiment are given the same reference numerals, and the description thereof will be omitted.

(1) Configuration of information processing system 3 FIG. 11 is a block diagram showing the configuration of information processing system 3 according to the third embodiment.

As shown in FIG. 11, the information processing system 2 includes an information processing device 10, a user terminal 20, a network 30, an accommodation facility 50, a first server 60, and a second server 70. The information processing device 10, the user terminal 20, the accommodation facility 50, the first server 60, and the second server 70 can communicate with each other via the network 30 using wired or wireless communication standards. It is connected to the.

The accommodation facility 50 is an accommodation facility provided to the user by a service provider. The accommodation facility 50 is, for example, a hotel, an inn, or the like. The accommodation facility 50 includes an information processing device 41 and a voice input device 42 installed in predetermined equipment. The predetermined equipment is, for example, a check-in terminal installed at the front desk of an accommodation facility.

The first server 60 is a travel agency server or a server that executes a web system provided by the travel agency. Note that the first server 60 may be a server of an accommodation facility or a server that executes a web system provided by the accommodation facility. Hereinafter, a case where the first server 60 is a server of a travel agency will be described as an example.

When the user signs a contract at a travel agency for a trip that includes at least the use of the accommodation facility 50, the travel agency acquires the user's first voice data. For example, a travel agency acquires the first voice data using a voice input device installed at the travel agency. Then, the person in charge at the travel agency transmits the acquired first voice data to the first server 60.

The first server 60 transmits the first audio data and information regarding the accommodation period to the information processing device 10 in response to a request from the information processing device 10 or automatically. Further, the first server 60 transmits user information and information regarding the use of the accommodation facility 50 to the second server 70 .

Note that if the first server 60 is a Web system, the first server 60 may acquire the user's first voice data from the user terminal 20.

The second server 70 is a server that manages the use of the accommodation facility 50. Specifically, the second server 70 manages the user's check-in status and the like. When the second server 70 receives information indicating that the user has checked in from the information processing device 10, it registers that the user has checked in.

(2) Functions of the information processing device 10

In this embodiment, the first audio data stored in the first data DB 121 is acquired before the user starts using the service. For example, the first audio data is acquired from the first server 60 at the timing when the user concludes an accommodation contract with a service provider via a travel agency.

The reception control unit 131 acquires the first audio data, information regarding the accommodation period, etc. from the first server.

The generation unit 134 generates a password valid for the received accommodation period.

The presenting unit 135 presents a password to the user using the information processing device 41 installed in a predetermined facility within the accommodation facility 50 when the user uses the service. Specifically, when the reception control unit 131 receives a password request from the information processing device 41, the presentation unit 135 transmits the password to the information processing device 41. Thereby, the presenting unit 135 presents the password to the user using an information processing device installed in a predetermined facility within the accommodation facility 50.

When the reception control unit 131 receives the second audio data from the information processing device 41, the authentication unit 136 performs user authentication using the first audio feature amount, the second audio feature amount, and the password. . In response to successful user authentication, the authentication unit 136 registers check-in by transmitting the authentication result of the user authentication to the second server. Specifically, when the user authentication is successful, the authentication unit 136 transmits the user authentication result and date and time to the second server 70 . Thereby, the authentication unit 136 causes the second server 70 to register that the user has checked in using the authentication result of the user authentication and the date and time.

(3) Operation In the following, processing in the information processing device 10 will be described with reference to the drawings.

FIG. 12 is a flowchart illustrating an example of the flow of authentication processing performed by the information processing device 10. The information processing device 10 executes this process at the timing when the password request is received from the information processing device 41, etc.

In step S602, the generation unit 134 generates a password valid for the received accommodation period.

In step S606, if the user authentication is successful, the authentication unit 136 registers the check-in by transmitting the authentication result of the user authentication to the second server, and ends the process.

(4) Summary According to the present disclosure, information regarding the accommodation period is acquired from the first server, and the password valid for the accommodation period is generated. Further, an information processing device installed in a predetermined facility within the accommodation facility presents a password to the user, and input of the second audio data is accepted through a voice input device installed in the facility. Then, in response to successful user authentication, check-in is registered by transmitting the authentication result of the user authentication to the second server. The service is the provision of accommodation facilities, and the first audio data is acquired from the first server before the user starts using the service. The first server is a server of an accommodation facility or a travel agency, or a web system provided by the accommodation facility or a travel agency, and the second server is a server that manages the use of the accommodation facility. As a result, the user can be authenticated without performing a process of learning the user's voice on the accommodation facility side. Moreover, the user can check in even if the accommodation facility does not have reception staff or the like. Therefore, the burden on service providers, such as labor shortages and measures to prevent infectious diseases, can be reduced.

Note that a physical key, electronic key, or password may be issued in response to registration of check-in. Specifically, the authentication unit 136 transmits the success of user authentication to the information processing device 41. Upon receiving successful user authentication, the information processing device 41 unlocks the locker storing the physical key, sends the electronic key to the user terminal 20, and presents the user with a password necessary for equipment in the facility. Provide the key to the user, such as by doing so. Furthermore, if a configuration is adopted in which voice authentication is performed in the same way as check-in when unlocking a locker, etc., authentication can be performed without issuing a physical key.

With such a configuration, a key related to the use of the accommodation facility is automatically issued through automatic check-in. Therefore, the service provider does not need to perform manual check-in, etc., and can solve the problem of labor shortage. Moreover, according to such a configuration, since no manual intervention is required, it is possible to prevent infectious diseases.

In addition, when physical keys are provided when family members or members of the same group use a facility, there are a limited number of holders of the physical keys, so actions within the facility may be influenced by the holders' actions. It ends up. However, the technology of the present disclosure can give each user unlocking authority linked to successful check-in authentication. Therefore, the technology of the present disclosure does not impose a physical burden on the accommodation facility and the users, and can improve the degree of freedom of movement of each user within the accommodation facility. Furthermore, the technology of the present disclosure can reduce the risk of physical loss such as loss or damage to a physical key.

Further, in the present embodiment, the second server 70 acquires information indicating that the user has checked in from the information processing device 10, but the present invention is not limited thereto. For example, the second server 70 may receive information indicating that the user has checked in from a terminal (for example, the information processing device 41) used for checking in to the accommodation facility 50.

Further, in this embodiment, the case of check-in has been described as an example, but of course it can also be used for check-out.

<Fourth embodiment>
In the fourth embodiment, an example will be described in which the above user authentication is used for authentication when using a service. In the fourth embodiment, a case will be described in which the work is performed by a service provider at a call center. The services include, for example, accepting repairs and inquiring about credit card usage details. Note that the same configurations as those in the first embodiment and the second embodiment are given the same reference numerals, and the description thereof will be omitted.

(1) Configuration of information processing system 4 FIG. 13 is a block diagram showing the configuration of information processing system 4 according to the fourth embodiment.

As shown in FIG. 13, the information processing system 2 includes an information processing device 10, a user terminal 20, a network 30, and a call center 80. The information processing device 10, the user terminal 20, and the call center 80 are connected to be able to communicate with each other via the network 30 using a wired or wireless communication standard.

The call center 80 includes an information processing device 81 operated by a person in charge. The information processing device 81 has the following functions.
- A function to talk to a user - A function to output a password to an output device (for example, a display) connected to the information processing device 81.
- A function of transmitting the second audio data acquired by the telephone call function to the information processing device 10.
- A function of outputting the fact that the authentication result of user authentication has been received from the information processing device 10 to an output device connected to the information processing device 81.

(2) Functions of the information processing device 10 The generation unit 134 generates a keyword used in a conversation as a password, and also generates a question to which the keyword is answered. Specifically, the generation unit 134 generates keywords that occur when a telephone receptionist has a conversation with a user. The generation unit 134 generates, for example, information regarding the user's personal information managed at the call center 80, climate information at the timing of the conversation, etc. as the keyword. The generation unit 134 also generates a question to which the generated keyword is answered. The generation unit 134 selects a question to which the generated keyword is answered by, for example, storing a keyword generation method and a question in advance in association with each other.

For example, assume that the method by which the generation unit 134 generates a keyword is a keyword included in the user's date of birth. Assume that the user's date of birth is, for example, April 1, 2000. In this case, the generation unit 134 generates the keyword as "nenshigatsu", which is a part of the user's date of birth. Then, the generation unit 134 generates a question such as "Please tell us your date of birth."

The presentation unit 135 presents a password and a question to be answered with a keyword to the person in charge of telephone reception of the person providing the service. Specifically, when the reception control unit 131 receives a password request from the information processing device 81, the presentation unit 135 transmits the password and the question to the information processing device 81. Thereby, the presentation unit 135 uses the information processing device 81 to present the password and question to the telephone receptionist. When the telephone receptionist asks the user a question, the information processing device 81 can receive the input of the second voice data.

When the reception control unit 131 receives the second audio data from the information processing device 81, the authentication unit 136 performs user authentication using the first audio feature amount, the second audio feature amount, and the password. . Then, the authentication unit 136 transmits the authentication result of the user authentication to the information processing device 81. Thereby, the information processing device 81 displays the authentication result to the person in charge of telephone reception, so that the person in charge of telephone reception can perform user authentication.

(3) Operation In the following, processing in the information processing device 10 will be described with reference to the drawings.

FIG. 14 is a flowchart illustrating an example of the flow of authentication processing performed by the information processing device 10. The information processing device 10 executes this process at the timing of receiving the password request from the information processing device 81, etc.

In step S701, the reception control unit 131 receives a password request from the information processing device 81.

In step S702, the generation unit 134 generates a keyword used in the conversation as a password, and also generates a question to which the keyword is answered.

In step S703, the presenting unit 135 presents a password and a question to be answered with a keyword to the person in charge of telephone reception of the person providing the service.

In step S704, the authentication unit 136 transmits the authentication result of the user authentication to the information processing device 81, and ends the process.

(4) Summary According to the present disclosure, keywords used in a conversation are generated as passwords, questions are generated to be answered by the keywords, and questions to be asked to be answered by the telephone receptionist of the person providing the service with the passwords and keywords. and present. Then, input of second voice data is accepted by asking a question to which a keyword is answered. This saves the trouble of transmitting personal information at the call center. Furthermore, even if the user's personal information is leaked, identity theft can be prevented. Furthermore, since the password is derived from the questions asked of the user, user authentication can be performed without the user being aware that he or she has been authenticated.
<Fifth embodiment>
In the fifth embodiment, an example will be described in which the above user authentication is used for authentication when using a service. In the fifth embodiment, a case will be described in which the service is the use of a delivery locker. Note that the same configurations as those in the first embodiment and the second embodiment are given the same reference numerals, and the description thereof will be omitted.

(1) Configuration of the information processing system 5 FIG. 15 is a block diagram showing the configuration of the information processing system 5 of the fifth embodiment.

As shown in FIG. 15, the information processing system 5 includes an information processing device 10, a user terminal 20, a network 30, a delivery person's terminal 90, and a delivery locker 91. The information processing device 10, the user terminal 20, and the delivery person's terminal 90 are connected to be able to communicate with each other via the network 30 using a wired or wireless communication standard.

The delivery person's terminal 90 is a mobile terminal operated by the delivery person. When the person in charge of delivery puts the package in the delivery locker, the person in charge of delivery inputs a character string into the terminal 90. The character string is, for example, a character string used as a password, a character string of information regarding luggage, or the like. The string of information regarding the package includes, for example, the address of the destination, the telephone number of the destination, the desired time for receiving the package, the date and time of delivery, and the management number. In the present disclosure, an example will be described in which a delivery person inputs a character string to be used as a password as is. The terminal 90 transmits a character string to the information processing device 10.

The delivery locker 91 includes an information processing device 41 and a voice input device 42 . The information processing device 41 has a function of unlocking the delivery locker 91 when the authentication result of user authentication is successful.

(2) Functions of the information processing device 10 The reception control unit 131 receives a temporary character string from the delivery person's terminal 90.

The generation unit 134 generates the received character string as a password. Note that if the character string is not used as it is as a password, the generation unit 134 generates a password from the character string using a predetermined method. The generation unit 134 generates a password by, for example, converting a character string into a readable character string using an arbitrary conversion method, calculating hash values of character strings, and combining the corresponding readable character strings. do.

The authentication unit 136 unlocks the locker in which the delivery item is stored by the delivery person in response to successful user authentication. Specifically, the authentication unit 136 transmits the authentication result of the user authentication to the delivery locker 91.

(3) Operation In the following, processing in the information processing device 10 will be described with reference to the drawings.

FIG. 16 is a flowchart illustrating an example of the flow of authentication processing performed by the information processing device 10. The information processing device 10 executes this process at an arbitrary timing (for example, at the timing when the user operates the delivery locker 91, etc.).

In step S802, the generation unit 134 generates the received character string as a password.

In step S806, the authentication unit 136 transmits the user authentication result to the delivery locker 91, and ends the process.

(4) Summary According to the present disclosure, a temporary character string is received from a delivery person's terminal, the received character string is generated as a password, and in response to successful user authentication, the delivery person Unlock the locker containing the delivered items. This eliminates the need for an absentee slip when a delivery person sets a password for a delivery locker.

Additionally, if there are multiple items to be delivered, multiple delivery lockers may be used. By setting one representative password for multiple delivery lockers, the user can unlock multiple delivery lockers at once with one authentication.
<Sixth embodiment>
In the sixth embodiment, an example will be described in which the above user authentication is used for authentication when using a service. In the sixth embodiment, a case where voice authentication is performed in a general locker will be described. Note that the same configurations as those in the first embodiment are given the same reference numerals and the description thereof will be omitted.

(1) Configuration of the information processing system 6 FIG. 17 is a block diagram showing the configuration of the information processing system 6 of the sixth embodiment.

As shown in FIG. 17, the information processing system 6 includes an information processing device 10, a user terminal 20, a network 30, and a locker 94. The information processing device 10, the user terminal 20, and the locker 94 are connected to be able to communicate with each other via the network 30 using a wired or wireless communication standard.

The locker 94 is connected to the information processing device 41 and the voice input device 42 .
The information processing device 41 further has the following functions.
- A function of unlocking the locker 94 in response to receiving an unlocking instruction from the information processing device 10.

(2) Functions of the information processing device 10 The reception control unit 131 receives an authentication request from the information processing device 41.

The generation unit 134 generates a password in response to receiving the authentication request.

The authentication unit 136 unlocks the locker used by the user in response to successful user authentication. Specifically, when the user authentication is successful, the authentication unit 136 transmits an instruction to unlock the locker to the information processing device 41. Thereby, the authentication unit 136 causes the information processing device 41 to unlock the locker in response to the unlocking instruction.

(3) Operation In the following, processing in the information processing device 10 will be described with reference to the drawings.

FIG. 18 is a flowchart illustrating an example of the flow of authentication processing performed by the information processing device 10. The information processing device 10 executes this process at the timing when the password request is received from the information processing device 41, etc.

In step S901, the reception control unit 131 receives an authentication request from the information processing device 41.

In step S902, the generation unit 134 generates a password in response to receiving the authentication request.

In step S906, if the user authentication is successful, the authentication unit 136 transmits an instruction to unlock the locker to the information processing device 41, and ends the process.

(4) Summary According to the present disclosure, an authentication request is received from another terminal, a password is generated in response to the acceptance of the authentication request, and a password is used by the user in response to successful user authentication. Execute unlocking of locker. This allows users to use lockers without using actual keys or image recognition.
For example, image recognition may be undesirable in lockers used at hot springs, pools, gyms, etc. due to the large amount of exposed skin and privacy concerns. Additionally, such lockers require the user to always carry a temporary physical key, which is cumbersome. For example, people had to do things like tie a locker key to their arm or ankle while in the bathhouse. According to the present disclosure, such privacy issues and user annoyance can be resolved.

<Modified example>
Although the disclosed embodiments have been described above, they can be implemented in various other forms, and can be implemented with various omissions, substitutions, and changes. These embodiments and modifications, as well as omissions, substitutions, and changes, are included within the technical scope of the claims and their equivalents.

For example, the generation unit 134 may be configured to generate a password on the condition that other authentications are successful. Other authentications include, for example, conventional password authentication, telephone number authentication, and other authentications other than voice authentication. In this case, the reception control unit 131 receives the fact that another authentication was successful. The generation unit 134 generates a password in response to success of other authentications. In this way, by combining with other authentications, the technology of the present disclosure can further increase security strength. Further, the authentication technology of the present disclosure can increase security strength by incorporating it into two-step authentication together with other authentications. In particular, in mobile terminals such as smartphones, when combined with fingerprint authentication or iris authentication, highly secure authentication can be performed without the user having to enter a character string.

Further, each function of the information processing device 10 may be configured in another device. For example, each DB of the storage unit 120 may be constructed as an external database. Further, each function of the information processing device 10 may be configured in another device. For example, each DB of the storage unit 120 may be constructed as an external database.

<Additional notes>
The matters described in each of the above embodiments are additionally described below.

(Additional Note 1) A program to be executed by a computer (for example, information processing device 10) including a processor (12) and a memory (11), the program being executed by a user registered in advance in the processor. a step of extracting a first voice feature amount from the first voice data (S304), a step of generating a temporary password (S402), a step of presenting the password to the user (S403), a step of receiving input of second voice data in which the user reads out the password (S404); a step of extracting a second voice feature amount from the received second voice data (S405); A step (S405) of performing user authentication using the feature amount, the second audio feature amount, and the password is executed, and the first audio feature amount and the second audio feature amount are used. is a program that is an audio feature expressed as a vector.

(Additional Note 2) In response to inputting voice data, further execute a step (S303) of acquiring a trained model that is trained in advance to output a voice feature amount representing a user's voice feature as a vector, and In the step of extracting a first voice feature, the first voice feature is extracted using the first voice data and the trained model, and in the step of extracting the second voice feature, , the program according to (Appendix 1), which extracts a second audio feature amount using the second audio data and the trained model.

(Additional Note 3) A step (S302) of converting the first audio data into a third audio feature represented by a two-dimensional vector using Mel frequency cepstrum coefficients; 2 audio data into a fourth audio feature represented by a two-dimensional vector (S461), and in the step of extracting the first audio feature, the third audio feature and the trained model, in the step of extracting a first audio feature amount, and extracting the second audio feature amount, using the fourth audio feature amount and the trained model. (Appendix 2), wherein the trained model is trained in advance to output the audio feature represented by the vector by inputting the two-dimensional vector. Programs listed.

(Additional Note 4) A step of converting the second audio data into text data (S464), a step of performing password authentication using the password and the text data (S465), and the first audio feature a step of performing speaker authentication (S466) according to the distance between the amount and the second voice feature amount, and in the step of performing user authentication, the authentication result of the password authentication and the speaker The program according to any one of (Appendix 1) to (Appendix 3), which performs user authentication using an authentication result of authentication.

In the step of performing user authentication, a step of permitting use of the service in response to successful user authentication is executed, and the first audio data is obtained before the user starts using the service. The program according to (Additional Note 1), wherein in the presenting step, the password is presented to the user when the user uses the service.

In the presenting step, the password is presented to the user by an information processing device installed in a predetermined equipment in the facility, and in the step of accepting the input of the second audio data, an information processing device installed in the equipment is installed in the equipment. receiving the input of the second audio data through the device, and in the step of allowing, unlocking the equipment or activating the device related to providing the service in response to the successful user authentication; The program described in (Appendix 5).

(Additional Note 7) Obtaining information regarding the accommodation period from the first server, and in the generating step, generating the password valid for the accommodation period, and in the presenting step, In the step of presenting the password to the user and accepting the input of the second audio data by an information processing device installed in a predetermined equipment in the facility, the second audio data is transmitted through the audio input device installed in the equipment. In the step of accepting the input of voice data and allowing it, in response to the success of the user authentication, registering that the user has checked in by transmitting the authentication result of the user authentication to the second server; The service is provision of the use of accommodation facilities, the first audio data is obtained from the first server before the user starts using the service, and the first voice data is obtained from the first server. , a server of the accommodation facility or travel agency, or a web system provided by the accommodation facility or travel agency, and the second server is a server that manages the use of the accommodation facility (Appendix 5) ).

(Additional note 8) In response to the registration of the check-in, execute a step of issuing a physical key, electronic key, or password for unlocking the facility or the device (Additional note 7) The program described in.

(Additional Note 9) The program according to (Additional Note 5), wherein in the presenting step, the password is presented to a telephone receptionist of the person providing the service.

(Additional note 10) In the step of generating, a keyword to be used in the conversation is generated as the password, and a question to be answered with the keyword is generated, and in the step of presenting the password, to the person in charge of telephone reception of the person providing the service. In the step of presenting the password and a question to which the keyword is to be answered, and receiving the input of the second voice data, receiving the input to the second voice data by asking the question to be answered by the keyword. 9) The program described in 9).

(Additional Note 11) In the step of receiving a temporary character string from the delivery person's terminal and the step of generating, the received character string is generated as the password, and in the step of authorizing, the user authentication is successful. The program according to (Appendix 5), wherein the delivery person unlocks the locker in which the delivery item is stored in response to the delivery.

(Additional Note 12) A step of receiving an authentication request from another terminal is executed, and in the step of generating, the password is generated in response to receiving the authentication request, and in the step of permitting, the user The program according to (Appendix 5), which unlocks a locker used by the user in response to successful authentication.

(Additional Note 13) The step of accepting that another authentication has been successful is executed, and in the generating step, the password is generated in response to the success of the other authentication. program.

(Additional Note 14) An information processing device (10) comprising a processor (12), the processor extracting a first voice feature amount from first voice data of a user registered in advance (S304); , a step of generating a temporary password (S402), a step of presenting the password to the user (S403), and a step of receiving input of second audio data in which the user reads out the password (S404). , a step of extracting a second audio feature amount from the received second audio data (S405), using the first audio feature amount, the second audio feature amount, and the password, The information processing apparatus executes a step of user authentication (S405), and the first audio feature amount and the second audio feature amount are audio features represented by vectors.

(Additional Note 15) A step (S304) in which a computer (for example, information processing device 10) including a processor (12) extracts a first voice feature amount from first voice data of a user registered in advance; a step of generating a password (S402), a step of presenting the password to the user (S403), a step of receiving input of second voice data in which the user reads out the password (S404); a step of extracting a second audio feature from the second audio data (S405); and performing user authentication using the first audio feature, the second audio feature, and the password. (S405), wherein the first audio feature amount and the second audio feature amount are audio features represented by vectors.

1 information processing system, 2 information processing system, 3 information processing system, 4 information processing system, 5 information processing system, 6 information processing system, 10 information processing device, 11 storage device, 12 processor, 13 input/output interface, 14 communication interface , 20 user terminal, 21 storage device, 22 processor, 23 input/output interface, 24 communication interface, 30 network, 40 facility, 41 information processing device, 42 voice input device, 50 accommodation facility, 60 first server, 70 second server, 80 call center, 81 information processing device, 90 terminal, 91 delivery locker, 94 locker, 110 communication unit, 120 storage unit, 130 control unit, 131 reception control unit, 132 transmission control unit, 133 extraction unit, 134 generation unit , 135 presentation section, 136 authentication section.

Claims (15)

A program for causing a computer including a processor and a memory to execute, the program causing the processor to:
extracting a first voice feature amount from first voice data of a user registered in advance;
generating a temporary password;
presenting the password to the user;
receiving input of second audio data in which the user reads out the password;
extracting a second audio feature from the received second audio data;
performing user authentication using the first audio feature amount, the second audio feature amount, and the password;
run the
The first audio feature amount and the second audio feature amount are audio features represented by vectors,
program. further performing the step of obtaining a trained model that has been trained in advance to output a voice feature amount representing a user's voice feature as a vector in response to inputting voice data;
In the step of extracting the first voice feature amount, extracting a first voice feature amount using the first voice data and the trained model,
In the step of extracting the second voice feature amount, extracting a second voice feature amount using the second voice data and the learned model;
The program according to claim 1. converting the first audio data into a third audio feature represented by a two-dimensional vector using Mel frequency cepstral coefficients;
converting the second audio data into a fourth audio feature represented by a two-dimensional vector using Mel frequency cepstral coefficients;
run the
In the step of extracting the first audio feature, extracting the first audio feature using the third audio feature and the trained model,
In the step of extracting the second voice feature, extracting a second voice feature using the fourth voice feature and the trained model,
The trained model is trained in advance to output the audio feature amount represented by the vector by inputting the two-dimensional vector.
The program according to claim 2. converting the second audio data into text data;
performing password authentication using the password and the text data;
performing speaker authentication according to the distance between the first audio feature amount and the second audio feature amount;
run the
in the step of performing user authentication, performing user authentication using the authentication result of the password authentication and the authentication result of the speaker authentication;
The program according to any one of claims 1 to 3. in the step of performing user authentication, allowing use of the service in response to successful user authentication;
run the
The first audio data is acquired before the user starts using the service,
In the presenting step, presenting the password to the user when the user uses the service;
The program according to claim 1. In the presenting step, the password is presented to the user by an information processing device installed in a predetermined equipment in the facility;
In the step of receiving the input of the second audio data, receiving the input of the second audio data through an audio input device installed in the equipment,
In the step of allowing, in response to the successful user authentication, unlocking the equipment or activating a device related to providing the service;
The program according to claim 5. obtaining information regarding the length of stay from the first server;
run the
In the step of generating, generating the password valid for the accommodation period,
In the presenting step, the password is presented to the user by an information processing device installed in a predetermined facility in the accommodation facility;
In the step of receiving the input of the second audio data, receiving the input of the second audio data through an audio input device installed in the equipment,
In the step of allowing, in response to the successful user authentication, registering check-in by transmitting the authentication result of the user authentication to a second server;
The service is the provision of accommodation facilities;
The first audio data is obtained from the first server before the user starts using the service,
The first server is a server of the accommodation facility or travel agency, or a web system provided by the accommodation facility or travel agency,
The second server is a server that manages the use of the accommodation facility,
The program according to claim 5. 8. The program according to claim 7, wherein the program executes the step of issuing a physical key, electronic key, or password for unlocking the facility or the device in response to the registration of the check-in. In the presenting step, presenting the password to a telephone receptionist of the person providing the service;
The program according to claim 5. In the step of generating, a keyword used in the conversation is generated as the password, and a question to be answered with the keyword is generated,
In the presenting step, presenting the password and a question for answering the keyword to a telephone receptionist of the person providing the service;
In the step of accepting input of the second voice data, receiving the input of the second voice data by asking a question to which the keyword is answered.
The program according to claim 9. receiving a temporary string from a delivery person's terminal;
In the step of generating, generating the received character string as the password,
In the step of allowing, in response to the successful user authentication, the delivery person unlocks the locker in which the delivery item is stored;
The program according to claim 5. a step of receiving an authentication request from another terminal;
run the
In the generating step, generating the password in response to receiving the authentication request,
In the step of allowing, in response to the successful user authentication, unlocking the locker used by the user;
The program according to claim 5. accepting that the other authentication was successful;
run the
In the generating step, generating the password in response to the success of the other authentication;
The program according to claim 1. An information processing device comprising a processor, the processor comprising:
extracting a first voice feature amount from first voice data of a user registered in advance;
generating a temporary password;
presenting the password to the user;
receiving input of second audio data in which the user reads out the password;
extracting a second audio feature from the received second audio data;
performing user authentication using the first audio feature amount, the second audio feature amount, and the password;
Run
The first audio feature amount and the second audio feature amount are audio features represented by vectors,
Information processing device. A computer equipped with a processor,
extracting a first voice feature amount from first voice data of a user registered in advance;
generating a temporary password;
presenting the password to the user;
receiving input of second audio data in which the user reads out the password;
extracting a second audio feature from the received second audio data;
performing user authentication using the first audio feature amount, the second audio feature amount, and the password;
Run
The first audio feature amount and the second audio feature amount are audio features represented by vectors,
Method.