Transaction data processing method and device and electronic equipment

Abstract

The embodiment of the invention discloses a transaction data processing method and device and electronic equipment, and the method comprises the steps: receiving a voice starting instruction which is initiated after a collection object constructs a transaction order on collection equipment; the biological recognition transaction equipment is in one-way communication with the collection equipment; extracting voiceprint features in the voice starting instruction, matching the voiceprint features in the voice starting instruction with stored target voiceprint features, and starting a biological recognition function according to the voice starting instruction if matching succeeds; collecting biological characteristics of the payment object by using a biological recognition function, and obtaining a payment account of the payment object based on the collected biological characteristics; and based on the payment account and the virtual resources required by the transaction in the transaction order, the collection account completes the payment transaction and closes the biological recognition function. By using the technical scheme provided by the invention, the probability of repeatedly processing data in biological recognition transaction payment is reduced, and the transaction completion efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a transaction data processing method, device and electronic device. Background Art

[0002] With the development of computer and Internet technology, online payment has gradually replaced cash payment. People can go shopping without carrying cash, which greatly facilitates people's lives. With the advancement of technology, there are more and more ways to pay online. The emergence of biometric payment devices allows people to complete payment transactions through identity verification through some biometric features such as fingerprints, voice, face, palm prints, etc., which is not only convenient but also safe.

[0003] Normally, when making biometric payment, a biometric transaction device is required to perform biometric identification to verify the user's identity. After the user's identity verification is passed, the user's account information is obtained, and then the payment is collected using a payment device (such as a POS (Point of Sales) machine) to complete the payment transaction. The communication method between the biometric transaction device and the payment device generally adopts a one-way communication mode, that is, simulating traditional keyboard input, and is fully compatible with all existing offline POS devices, that is, it can be accessed by scanning a barcode gun. However, in this one-way communication mode, the biometric transaction device can only send information to the payment device in one direction, and the payment device cannot send information to the biometric transaction device, which may cause the biometric transaction device to be unable to perceive the status information of the payment device, such as whether the conditions for biometric payment are currently met. If the biometric feature is collected without meeting the conditions, the current biometric transaction device will enter the order checking state, but this payment will fail, resulting in a transaction failure, and the user needs to re-perform biometric payment, which affects the efficiency of the transaction and the physical experience is also relatively poor. Biometric transaction devices and payment devices generally belong to different managers and producers respectively. To achieve two-way communication between biometric transaction devices and payment devices, both parties need to cooperate in technical transformation of the equipment, which is very unfriendly to the rapid adaptation of offline stock devices, and there are many compatibility issues and relatively many technical difficulties.

[0004] Therefore, how to provide a transaction data processing solution that can reduce the probability of duplicate transaction data processing and improve transaction data processing efficiency during biometric payment transactions is a technical problem that urgently needs to be solved in this field. Summary of the invention

[0005] The embodiments of this specification provide a transaction data processing method, device, equipment, storage medium, and computer program product, which reduce the probability of duplication in the biometric payment transaction processing process and improve the efficiency of transaction data processing.

[0006] On the one hand, an embodiment of this specification provides a transaction data processing method, which is applied to a biometric transaction device, and the biometric transaction device is used to collect biometric features during the transaction data processing process. The method includes:

[0007] receiving a voice start instruction, the voice start instruction being initiated by the payment recipient after the payment device constructs a transaction order, the voice start instruction being used to start the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicating unidirectionally with the payment device;

[0008] Extracting the voiceprint feature in the voice start instruction, matching the voiceprint feature in the voice start instruction with the stored target voiceprint feature, and if the match is successful, starting the biometric recognition function according to the voice start instruction;

[0009] Using the biometric recognition function to collect biometric features of the payment object, and obtaining the payment account of the payment object based on the collected biometric features;

[0010] Based on the payment account and the virtual resources required for the transaction in the transaction order, the payment transaction is completed and the biometric recognition function is turned off.

[0011] On the other hand, a transaction data processing device is provided, the device is applied to a biometric transaction device, the biometric transaction device is used to collect biometric features during transaction data processing, the device comprises:

[0012] A voice start instruction receiving module, used for receiving a voice start instruction, wherein the voice start instruction is initiated by the payment recipient after the payment device constructs a transaction order, and the voice start instruction is used for starting the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicates unidirectionally with the payment device;

[0013] A voiceprint recognition module, used to extract the voiceprint features in the voice start instruction, match the voiceprint features in the voice start instruction with the stored target voiceprint features, and if the match is successful, start the biometric recognition function according to the voice start instruction;

[0014] A biometric feature recognition module, used to collect the biometric features of the payment object using the biometric recognition function, and obtain the payment account of the payment object based on the collected biometric features;

[0015] The transaction processing module is used to complete the payment transaction based on the payment account and the virtual resources required for the transaction in the transaction order, and to close the biometric recognition function.

[0016] Furthermore, the voice start instruction receiving module is specifically used for:

[0017] Using an array microphone provided in the biometric transaction device to collect voice data, and determining the sound source position and direction corresponding to the collected voice data according to the array microphone;

[0018] Based on the sound source position and direction corresponding to the voice data, determining the target voice data belonging to the location of the payment object from the voice data;

[0019] The target voice data is used as the voice start instruction.

[0020] Furthermore, the voiceprint recognition module is specifically used for:

[0021] Extracting a voiceprint feature vector corresponding to each frame of the voice signal in the voice start instruction;

[0022] The extracted voiceprint feature vectors are concatenated in chronological order to obtain a voiceprint feature sequence;

[0023] The voiceprint feature sequence is input into a pre-trained voiceprint recognition model, and the voiceprint recognition model is used to calculate the matching probability between the voiceprint feature sequence and each of the target voiceprint features. If the matching probability between the voiceprint feature sequence and at least one target voiceprint feature is greater than a preset threshold, the match is determined to be successful, and the biometric recognition function is started according to the voice start instruction; wherein the voiceprint recognition model is constructed based on the training of the target voiceprint features.

[0024] Furthermore, the device also includes a voiceprint recognition model building module, which is used to:

[0025] Extracting a sample voiceprint feature vector from each frame of voice signal in each voice sample data; wherein each voice sample data is voice data of a target object that is allowed to activate the biometric recognition function of the biometric recognition transaction device;

[0026] The sample voiceprint feature vectors corresponding to each voice sample data are concatenated in chronological order to obtain a sample voiceprint feature sequence corresponding to each voice sample data, and the sample voiceprint feature sequence is used as the target voiceprint feature;

[0027] The probability distribution of the target voiceprint feature is extracted using a Gaussian mixture model, and the voiceprint recognition model is trained and constructed; wherein the voiceprint recognition model includes multiple voiceprint recognition sub-models, and each voiceprint recognition sub-model corresponds to a Gaussian distribution of the target voiceprint feature.

[0028] Furthermore, the device also includes a voiceprint registration module for:

[0029] Before receiving the voice start instruction, receiving a voiceprint registration request initiated by the payment recipient, the voiceprint registration request including: the voice sample data of the payment recipient, the voice start instruction and the instruction function corresponding to the voice start instruction, the instruction function is to start the biometric recognition function of the biometric recognition transaction device;

[0030] Extracting target voiceprint features from the voice sample data and storing the target voiceprint features;

[0031] The voice activation instruction is associated with an operation corresponding to an instruction function for activating a biometric function of the biometric transaction device.

[0032] Furthermore, the device also includes a transaction query module, which is used to:

[0033] receiving a voice query instruction, wherein the voice query instruction includes: a query time;

[0034] The voiceprint feature in the voice query instruction is extracted, and the voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information of the biometric transaction device within the query time is queried according to the voice query instruction.

[0035] Furthermore, the voice query instruction also includes the identity of the payment recipient to be queried, and the transaction query module is further used to:

[0036] The voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information of the identity identifier within the query time in the biometric transaction device is queried according to the voice query instruction.

[0037] On the other hand, a biometric transaction device for transaction data processing is provided, the device comprising:

[0038] A voice collection module, used for collecting voice start instructions; the voice start instructions are initiated by the payment recipient after the payment device constructs a transaction order, and the voice start instructions are used to start the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicates unidirectionally with the payment device;

[0039] Voiceprint command module, used for voiceprint registration and the configuration of the association between voice start commands and command functions;

[0040] A biometric feature collection module, used for collecting biometric features;

[0041] a transaction service module, configured to perform voiceprint recognition on the voice start instruction collected by the voice collection device, match the voiceprint feature in the voice start instruction with the target voiceprint feature, and after the biometric recognition function is turned on, perform biometric recognition based on the biometric feature collected by the biometric feature collection module, obtain the payment account of the payment object corresponding to the biometric feature, and complete the payment transaction based on the payment account and the virtual resources required for the transaction in the transaction order;

[0042] The state control module is used to enable the biometric recognition function based on the instruction function associated with the voice start instruction after the transaction service module determines that the voiceprint feature in the voice start instruction is successfully matched.

[0043] Another aspect provides an electronic device, including: a processor;

[0044] a memory for storing instructions executable by the processor;

[0045] The processor is configured to execute the instructions to implement any of the transaction data processing methods described above.

[0046] On the other hand, a computer-readable storage medium is provided. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute any of the above-mentioned transaction data processing methods.

[0047] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the transaction data processing method provided in the above various optional implementations.

[0048] The transaction data processing method, device, equipment, storage medium, and computer program product provided in the embodiments of this specification have the following technical effects:

[0049] The transaction data processing method provided in the embodiment of this specification is that after the payment device completes the transaction order, the payment object initiates a voice start command to the biometric transaction device. The biometric transaction device extracts and identifies the voiceprint features in the voice start command, and matches the voiceprint features in the voice start command with the target voiceprint features collected and stored in advance to identify whether the identity of the payment object can control the biometric transaction device. After the voiceprint feature is successfully matched, the biometric recognition function of the biometric transaction device is turned on to collect the biometric features of the payment object, and the payment account of the payment object is obtained based on the collected biometric features, and then the payment transaction is completed according to the payment account and the transaction order. After the payment transaction is completed, the biometric recognition function of the biometric transaction device is turned off, so that the biometric recognition function of the biometric transaction device is turned on only after the payment device completes the creation of a transaction order, avoiding invalid collection of biometric features and improving the effective collection rate of biometric features. In this way, the transaction is completed once, and the biometric features will not be scanned repeatedly, which reduces the probability of repeated processing of transaction data during biometric payment transactions and improves the efficiency of transaction data processing. At other times, the biometric function is turned off. Even if the user scans the biometrics, it will not affect the processing of the transaction order, thus avoiding the problem of transaction failure and invalid collection of biometrics due to incorrect biometric scanning time. In addition, by triggering the start of the biometric function through voice, there is no need for two-way communication between the payment device and the biometric transaction device. That is, when the biometric transaction device communicates with the payment device in one direction, the payment device does not need to cooperate in technical transformation, and the transaction data of the existing biometric transaction devices can be completed accurately and quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions and advantages in the embodiments of this specification or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 What is shown is merely a schematic diagram of an application environment of a transaction data processing method;

[0052] Figure 2 It is a flowchart of a transaction data processing method provided by an embodiment of this specification;

[0053] Figure 3 is a schematic diagram of the structure of a biometric transaction device for transaction data processing in one embodiment of the present specification;

[0054] Figure 4This is a data interaction diagram of the process of starting the palm payment function of the palm transaction device in a scenario example of this specification;

[0055] Figure 5 This is a data interaction diagram of the entire palm payment process in a scenario example in this manual;

[0056] Figure 6 It is a structural schematic diagram of a transaction data processing device provided in an embodiment of this specification;

[0057] Figure 7 is a block diagram of an electronic device for transaction data processing provided by an embodiment of this specification;

[0058] Figure 8 It is a block diagram of another electronic device for transaction data processing provided in an embodiment of this specification. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this specification.

[0060] It should be noted that the terms "first", "second", etc. in the description and claims of the embodiments of this specification and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of this specification described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0061] See also Figure 1 , Figure 1 It is a schematic diagram of an application environment of a transaction data processing method provided in an embodiment of this specification, and the application environment may at least include a server 100 and a terminal 200.

[0062] In an optional embodiment, the server 100 can be used to receive the biometric features sent by the terminal 200, identify the payment object based on the biometric features, and complete the payment transaction. The server 100 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0063] In an optional embodiment, the terminal 200 may be a terminal that sends biometrics to the server 100, and may also perform voiceprint recognition and switch working states locally at the terminal, or may perform identity recognition of the payment object and payment transaction processing based on biometrics locally at the terminal as needed. Specifically, the terminal 200 may be a device with a biometric collection function, a device specifically used for biometric collection and assisting in payment transactions, or a device that adds biometric collection to some smart devices and can perform biometric payment transactions, such as: a smart phone, desktop computer, tablet computer, laptop computer and other electronic devices with a biometric collection function; or software running on the above electronic devices, such as an application, applets, etc. The operating system running on the electronic device in the embodiments of this specification may include but is not limited to Android, IOS, Linux, Windows, etc.

[0064] In addition, it should be noted that Figure 1 What is shown is merely a schematic diagram of an application environment of a transaction data processing method, and the embodiments of this specification are not limited to the above.

[0065] In the embodiments of this specification, the server 100 and the terminal 200 may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiments of this specification.

[0066] In the transaction data processing method of the embodiment of this specification, after the payment device completes the construction of the transaction order, the payment object can initiate a voice start command to control the biometric transaction device to start the biometric recognition function, collect the biometric characteristics of the payment object, and then complete the payment transaction. This avoids the problem that the payment device starts collecting biometric characteristics before completing the construction of the transaction order, resulting in the failure of the payment transaction and the need to re-collect biometric characteristics, reduces the probability of repeated processing of transaction data during biometric payment transactions, and improves the efficiency of completing payment transactions.

[0067] The following describes a transaction data processing method according to an embodiment of the present specification. Figure 2It is a flowchart of a transaction data processing method provided by an embodiment of this specification. This specification provides method operation steps such as the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only execution order. When the actual system or server product is executed, it can be executed in sequence or in parallel according to the method shown in the embodiment or the accompanying drawings (for example, in a parallel processor or multi-threaded processing environment). The method can be applied to terminal devices such as computers, tablet computers, and smart phones. Of course, it can also be applied to servers according to actual needs. The embodiments of this specification do not specifically limit this. Specifically, Figure 2 As shown, the method can be applied to biometric transaction equipment, which can be used to collect biometric features during the transaction data processing process, so as to identify the identity information of the payment object based on the biometric features, and then obtain the payment account of the payment object to complete the payment transaction. Among them, the biometric transaction equipment can be used for face recognition, fingerprint recognition, palm recognition, voice recognition, etc., and this specification embodiment does not specifically limit this. Figure 2 As shown, the method may include:

[0068] S202: Receive a voice start instruction, which is initiated by the payment recipient after the payment device constructs a transaction order. The voice start instruction is used to start the biometric function of the biometric transaction device; the payment device communicates unidirectionally with the biometric transaction device.

[0069] In the specific implementation process, when a user needs to make a payment transaction after shopping, the payment recipient (such as a cashier) can construct a transaction order by operating the payment recipient device. The transaction order may include information related to the transaction, such as relevant information of the two parties to the transaction and relevant information of the transaction recipient. Specifically, the transaction order may include the name of the payment recipient, the account, the name of the transaction commodity, the price, the transaction time, etc. The payment recipient device may be a device that can construct a transaction order and complete the transaction collection, such as a POS (Point of Sales) machine or other cash register device in a shopping mall or supermarket. The embodiment of this specification does not specifically limit the type of the payment recipient device. When the payment recipient device constructs the transaction order, it means that the payment operation can be performed. The payment recipient can initiate a voice start command to the biometric transaction device, and the voice start command can be used to start the biometric function of the biometric transaction device. That is to say, under normal circumstances, the biometric function of the biometric transaction device is turned off. When the payment recipient turns on the biometric function of the biometric transaction device through a voice start command, the biometric transaction device can start the function of collecting and identifying biometric features. The specific content of the voice start instruction can be set according to actual needs, such as: "Please scan your palm", "Please scan your face", "Start the scanning function", etc. The embodiments of this specification do not specifically limit the specific content of the voice start instruction.

[0070] In addition, the payment device and the biometric transaction device in the embodiment of this specification are one-way communication, that is, the biometric transaction device can send information to the payment device, but the payment device cannot send information to the biometric device. Generally, the payment device and the biometric transaction device in the embodiment of this specification belong to different objects, and the objects can be understood as the owner, manager, producer or developer of the device. The object of the payment device is generally a merchant such as a supermarket, while the object of the biometric transaction device is generally the platform corresponding to the payment account of the payment object, such as a financial institution or a mobile payment platform. Due to the difference in the objects, there are some technical barriers between the payment device and the biometric transaction device, and generally two-way communication is not possible. To achieve two-way communication, both parties need to cooperate in technical transformation, which is difficult to operate. Therefore, when the payment device has not completed the construction of the transaction order, the biometric transaction device collects biometric features, resulting in the failure of the payment transaction. Therefore, it is necessary to re-construct the order, collect biometric features, and repeat the processing of transaction data.

[0071] S204: extracting the voiceprint features in the voice start instruction, matching the voiceprint features in the voice start instruction with the stored target voiceprint features, and if the match is successful, starting the biometric recognition function according to the voice start instruction.

[0072] In the specific implementation process, after receiving the voice start command initiated by the payment recipient, the voiceprint feature in the voice start command can be extracted, and the extracted voiceprint feature can be matched with the pre-stored target voiceprint feature. If the match is successful, it is determined that the payment recipient who currently initiates the voice start command is an object that allows the use of voice start commands to control the biometric transaction device, and then the biometric function of the biometric transaction device can be activated according to the voice start command. Among them, the target voiceprint feature can be the voiceprint feature of the user who is allowed to use the voice start command to control the biometric transaction device to activate the biometric function. The voiceprint feature of the user who is allowed to control the biometric transaction device can be collected in advance as the target voiceprint feature and stored in the biometric transaction device.

[0073] For example, a supermarket adopts biometric payment technology, which can collect the voiceprint features of supermarket cashiers A and B in advance, such as collecting the voiceprint features of cashiers A and B initiating voice activation commands to control the biometric transaction device to activate the biometric function, and storing the collected voiceprint features in the biometric transaction device. When cashier A or B is cashing, after constructing a transaction order on the payment device, cashier A or B can control the biometric transaction device to activate the biometric function through voice activation commands.

[0074] Of course, before extracting the voiceprint features, the voice signal of the received voice start command may be preprocessed by noise reduction to improve the accuracy of voiceprint feature extraction.

[0075] S206: Collect the biometric features of the payment object using the biometric recognition function, and obtain the payment account of the payment object based on the collected biometric features.

[0076] In the specific implementation process, after the biometric recognition function is turned on, the biometric recognition transaction device enters a recognizable state, and then the biometric characteristics of the payment object can be collected. Among them, the payment object can be understood as the payer corresponding to the current transaction order, and the biometric characteristics of the payment object can be face, voice, fingerprint, palm print, etc., which can be set according to actual needs, and the embodiments of this specification do not make specific limitations. After the biometric characteristics of the payment object are collected, the payment account of the payment object can be obtained based on the biometric characteristics. Among them, the payment account can be the account information or payment code associated with the biometric characteristics of the payment object, and the specific content of the payment account can be adjusted according to actual needs, and the embodiments of this specification do not make specific limitations. In general, users who choose to use biometric recognition for transaction payment have already bound their biometric characteristics to a specific account in advance, so the corresponding account information can be obtained through the biometric characteristics.

[0077] In addition, in some embodiments, when the biometric transaction device turns on the biometric function, it can send prompt information to the payment object, such as: the light in the recognition area of ​​the biometric transaction device flashes, the light changes, or the biometric transaction device directly sends a voice reminder, etc., thereby facilitating the payment object to scan the biometric characteristics.

[0078] S208: Based on the payment account and the virtual resources required for the transaction in the transaction order, the payment transaction is completed and the biometric recognition function is turned off.

[0079] In the specific implementation process, after obtaining the payment account of the payment object, the payment transaction can be completed according to the payment account and the virtual resources required for the transaction in the transaction order. For example, the payment account can be transferred to the receiving account according to the virtual resources required for the transaction, or the payment account can be returned to the receiving device, and the receiving device will collect the payment from the payment account according to the virtual resources required for the transaction. In addition, after the payment transaction is completed, the biometric recognition function can be turned off to avoid other users scanning the biometric features in advance when the receiving device has not completed the construction of the transaction order, resulting in the failure of other transactions. At the same time, the biometric recognition function does not need to be turned on all the time, which avoids the collection of invalid biometric features and reduces the waste of resources.

[0080] For example, if user S goes shopping in a supermarket, he chooses to pay by palm swiping when making payment at the checkout counter. After cashier B settles the goods purchased by user S, he constructs a transaction order for the current transaction through the payment device. The transaction order may include the supermarket's payment account and the virtual resource 100 yuan required for the transaction. After the payment device constructs the transaction order, cashier B issues a voice start instruction, such as "Please swipe your palm". After receiving the voice start instruction, the biometric transaction device extracts the voice print feature in the voice start instruction and matches it with the pre-stored target voice print feature. After the match is successful, it confirms that the identity of cashier B is allowed to control the biometric transaction device. The biometric transaction device turns on the biometric function based on the voice start instruction. User S puts his palm into the collection area of ​​the biometric transaction device. The biometric transaction device collects the biometric feature of user S, that is, the palm print feature, and obtains the payment account of user S based on the collected biometric feature. The biometric transaction device can transfer 100 yuan from the payment account of user S to the collection account of the supermarket. The biometric transaction device can also send the payment account of user S to the collection device, and the collection device will process the payment. For example, the biometric transaction device can return the payment code of user S to the collection device, and the collection device transfers 100 yuan from the account of user S to its own collection account based on the payment code. After the payment transaction is completed, the biometric function of the biometric transaction device is turned off. At this time, even if the user swipes his palm, the user's biometric features will not be collected, and the data processing process of other transactions will not be affected.

[0081] A transaction data processing method provided in an embodiment of the present specification is that after the payment device completes the construction of the transaction order, the payment object initiates a voice start command to the biometric transaction device, and the biometric transaction device extracts and identifies the voiceprint features in the voice start command, and matches the voiceprint features in the voice start command with the target voiceprint features collected and stored in advance to identify whether the identity of the payment object can control the biometric transaction device. After the voiceprint feature is successfully matched, the biometric recognition function of the biometric transaction device is turned on to collect the biometric features of the payment object, and the payment account of the payment object is obtained based on the collected biometric features, and then the payment transaction is completed according to the payment account and the transaction order. After the payment transaction is completed, the biometric recognition function of the biometric transaction device is turned off, so that the biometric recognition function of the biometric transaction device is turned on only after the payment device completes the creation of a transaction order, avoiding invalid collection of biometric features and improving the effective collection rate of biometric features. In this way, the transaction is completed once, and there is no need to scan the biometric features repeatedly, which reduces the probability of repeated processing of transaction data during biometric payment transactions and improves the efficiency of transaction completion. At other times, the biometric function is turned off, and even if the user scans the biometrics, it will not affect the processing of the transaction order, thus avoiding the problem of transaction failure and invalid collection of biometrics due to incorrect biometric scanning time. In addition, by triggering the start of the biometric function through voice, there is no need for two-way communication between the payment device and the biometric transaction device, that is, when the biometric transaction device communicates with the payment device in one direction, there is no need for the payment device to cooperate in technical transformation, and the transaction data processing efficiency of the existing biometric transaction devices can be improved.

[0082] In some embodiments of this specification, the receiving of a voice activation instruction includes:

[0083] The array microphone provided in the biometric transaction device is used to collect voice data, and the sound source position and direction corresponding to the collected voice data are determined according to the array microphone;

[0084] Based on the sound source position and direction corresponding to the voice data, target voice data belonging to the location of the payment object is determined from the voice data;

[0085] The target voice data is used as a voice start instruction.

[0086] In the specific implementation process, in the embodiments of this specification, an array microphone can be set in the biometric transaction device, and the array microphone can be used to collect voice data. Microphone array refers to the arrangement of microphones, generally composed of a certain number of acoustic sensors (generally microphones), and is a system used to sample and process the spatial characteristics of the sound field. The array microphone can be used to collect voice data in the environment, and the phase and amplitude differences between the sound waveforms received by each microphone in the array microphone are used to analyze and calculate through a signal processing algorithm to determine the sound source position and direction corresponding to the collected voice data, thereby achieving directional collection of the sound source. Based on the determined sound source position and direction corresponding to the voice data, it is determined whether there is voice data at the location of the payment object of the payment device in the voice data. If so, the voice data at the location of the payment object is used as the target voice data, and the target voice data is used as the voice start instruction.

[0087] Normally, the recipient receives payment at the location of the payment device, and the biometric transaction device is generally fixed near the payment device. Therefore, the location of the recipient relative to the biometric transaction device is fixed. The array microphone is used to collect and identify the sound source position and direction of the voice data, so as to locate which voice data is issued by the recipient. The voice start command issued by the recipient can be accurately collected, and then the biometric function of the biometric transaction device can be activated, thereby improving the accuracy of the activation of the biometric function of the biometric transaction device. Especially in some noisy scenes, there may be a lot of voice data in the environment. The array microphone can improve the accuracy of voice start command collection, improve the signal-to-noise ratio and clarity of the voice signal, and reduce the interference of environmental noise.

[0088] In some embodiments, extracting the voiceprint feature in the voice start instruction, matching the voiceprint feature in the voice start instruction with the stored target voiceprint feature, and if the match is successful, starting the biometric recognition function according to the voice start instruction, includes:

[0089] Extract the voiceprint feature vector corresponding to each frame of voice signal in the voice start command;

[0090] The extracted voiceprint feature vectors are concatenated in chronological order to obtain a voiceprint feature sequence;

[0091] The voiceprint feature sequence is input into a pre-trained voiceprint recognition model, and the voiceprint recognition model is used to calculate the matching probability between the voiceprint feature sequence and each target voiceprint feature. If the matching probability between the voiceprint feature sequence and at least one target voiceprint feature is greater than a preset threshold, the match is determined to be successful, and the biometric recognition function is started according to the voice start instruction; wherein the voiceprint recognition model is constructed based on the target voiceprint feature training.

[0092] In the specific implementation process, the target voiceprint feature training can be used to build a voiceprint recognition model, and the trained voiceprint recognition model can be stored in the biometric transaction device. After receiving the voice start instruction, the voice start instruction can be divided into multiple voice frames, and the voiceprint feature vector corresponding to each frame of the voice signal can be extracted, and then the extracted voiceprint feature vectors are spliced ​​in chronological order to obtain a voiceprint feature sequence. The obtained voiceprint feature sequence is input into the voiceprint recognition model, and the voiceprint recognition model can automatically calculate the matching probability between the voiceprint feature sequence and the target voiceprint feature of the training model. If the matching probability of the voiceprint feature sequence and at least one target voiceprint feature is greater than the preset threshold, it is determined that the voiceprint feature in the voice start instruction successfully matches the at least one target voiceprint feature, and the target voiceprint feature with the largest matching probability can be used as the target voiceprint feature that matches the voiceprint feature in the voice start instruction, and the biometric recognition function is started according to the voice start instruction. Among them, the value of the preset threshold can be set according to actual needs, and the embodiments of this specification are not specifically limited.

[0093] For example: the target voiceprint features of cashiers A and B are used as sample data in advance to train and construct a voiceprint recognition model, and the voiceprint recognition model is stored in the biometric transaction device. When a voice start command is received, the voiceprint feature vector and voiceprint feature sequence in the voice start command are extracted, and the voiceprint feature sequence is input into the trained voiceprint recognition model, and the voiceprint recognition model calculates the matching probability between the input voiceprint feature sequence and the target voiceprint features of cashiers A and B. If it is determined through calculation that the matching probability between the input voiceprint feature sequence and the target voiceprint feature of cashier A is greater than a preset threshold, it is determined that the voiceprint feature of the voice start command matches the target voiceprint feature of cashier A successfully, that is, the voice start command is initiated by cashier A. At this time, the biometric recognition function of the biometric transaction device can be enabled based on the voice start command.

[0094] The embodiments of this specification first build a voiceprint recognition model based on the target voiceprint feature training that allows the control of the biometric function of the biometric transaction device. After receiving the voice start command, the voiceprint recognition model is used to calculate the matching probability between the voiceprint features in the voice start command and the target voiceprint features of the training model, thereby achieving identity authentication of the user who initiated the voice start command, ensuring the accurate activation of the biometric function of the biometric transaction device, and facilitating data processing of subsequent payment transactions.

[0095] In some embodiments of this specification, the method for constructing a voiceprint recognition model includes:

[0096] Extracting a sample voiceprint feature vector from each frame of voice signal in each voice sample data; wherein each voice sample data is voice data of a target object that is allowed to enable the biometric recognition function of the biometric recognition transaction device;

[0097] The sample voiceprint feature vectors corresponding to each speech sample data are concatenated in chronological order to obtain the sample voiceprint feature sequence corresponding to each speech sample data, and the sample voiceprint feature sequence is used as the target voiceprint feature;

[0098] A Gaussian mixture model is used to extract the probability distribution of the target voiceprint features, and a voiceprint recognition model is trained and constructed; wherein the voiceprint recognition model includes multiple voiceprint recognition sub-models, and each voiceprint recognition sub-model corresponds to a Gaussian distribution of the target voiceprint features.

[0099] In the specific implementation process, the voice data of the target object that allows the biometric recognition function of the biometric transaction device to be enabled can be first collected as voice sample data. The voice sample data can include the voice data of the voice start command. Of course, other voice data of the target object can also be collected as voice sample data when training the voiceprint recognition model. Extract the sample voiceprint feature vector of each frame of the voice signal in each voice sample data, splice the extracted sample voiceprint feature vectors in the order of time, obtain the sample voiceprint feature sequence corresponding to each voice sample data, and use the sample voiceprint feature sequence as the target voiceprint feature. Then use the Gaussian mixture model to extract the probability distribution of the target voiceprint feature, and train to construct a voiceprint recognition model. Among them, the Gaussian mixture model can be understood as a probabilistic statistical model. In the embodiment of this specification, the voiceprint recognition model constructed based on the Gaussian mixture model training can include multiple voiceprint recognition sub-models, and each voiceprint recognition sub-model can correspond to a Gaussian distribution of a target voiceprint feature. Then, when performing voiceprint recognition in the subsequent voiceprint recognition, the matching probability between the voiceprint feature in the voice start command and each voiceprint recognition sub-model, that is, the matching probability between each target voiceprint feature, can be calculated. The embodiments of this specification utilize Gaussian model training to construct a voiceprint recognition model, which facilitates the subsequent matching of voiceprint features and improves the efficiency of transaction data processing.

[0100] In addition, in some embodiments of this specification, not only can the biometric transaction device be controlled by voice to start the biometric function, but other voice commands (such as voice query commands) can also be used to control the biometric transaction device to perform other operations. Therefore, when training the voiceprint recognition model, voice sample data containing different voice commands can be collected to train and construct a voiceprint recognition model that can recognize the voiceprint features of different voice commands. The training method of the voiceprint features of other voice commands is similar to the training method of the voiceprint features corresponding to the voice start command, and will not be repeated here.

[0101] In some embodiments of the present specification, before receiving the voice start instruction, the method further includes:

[0102] Receiving a voiceprint registration request initiated by a payment recipient, the voiceprint registration request including: voice sample data of the payment recipient, a voice start instruction, and an instruction function corresponding to the voice start instruction, wherein the instruction function is to start a biometric function of the biometric transaction device;

[0103] Extracting target voiceprint features from the speech sample data and storing the target voiceprint features;

[0104] The voice activation instruction is associated with an operation corresponding to the instruction function of activating the biometric function of the biometric transaction device.

[0105] In the specific implementation process, before receiving the voice start instruction, voiceprint registration can be performed, and a voiceprint registration request can be received. The voiceprint registration request can include voice sample data, voice instructions, and the instruction function corresponding to the voice instruction. The voiceprint registration request is generally initiated by the user who is allowed to control the biometric transaction device, such as: the recipient. The voice instruction can include the voice start instruction in the above embodiment. The instruction function can be understood as the operation of the biometric transaction device by the voice instruction corresponding to the voice sample data, such as: turning on the biometric function or querying the transaction record of the biometric transaction device, etc. The instruction function can be set by the user who registered the voiceprint. After receiving the voice registration request, the target voiceprint feature in the voice sample data can be extracted, and the extracted target voiceprint feature can be stored. An identity identifier can also be configured for each target voiceprint feature. The identity identifier is used to mark the identity information of the recipient corresponding to the target voiceprint feature, so as to facilitate the subsequent identity authentication of the user who initiated the voice instruction when the biometric transaction device is controlled by voice.

[0106] At the same time, the voice command in the voiceprint registration can also be associated with the command function, such as: the voice start command is "please swipe your palm", and the corresponding command function is to start the biometric function of the biometric transaction device. When registering the voiceprint, the voice start command "please swipe your palm" can be associated with the biometric function of starting the biometric transaction device. Among them, the voice command can be the command corresponding to the voice signal in the voice sample data, or it can be a voice command specially recorded by the payment recipient. When registering the voiceprint, the voice data of the payment recipient saying "please swipe your palm" can be recorded as the voice sample data, and the command function corresponding to the voice command is set to start the biometric function of the biometric transaction device, then the voice command "please swipe your palm" can be associated with the biometric function of starting the biometric transaction device. When the voice command "please swipe your palm" is received, after the voiceprint recognition match is passed, the operation associated with the voice command can be executed to start the biometric function of the biometric transaction device. Alternatively, when registering the voiceprint, the voice data of the payment recipient speaking different voices can be recorded as the voice sample data to extract the voiceprint features of the payment recipient as the target voiceprint features. Then perform the voice command association operation, separately record the voice data of the payment recipient saying the specified voice command, and set the command function that the voice command needs to be associated with, identify whether the voiceprint feature corresponding to the voice command matches the voiceprint feature of the payment recipient, if it matches, associate the voice command with the operation corresponding to the set command function.

[0107] Of course, voice commands may include voice start commands, voice query commands, etc. Different voice commands may correspond to different command functions. Other voice commands may be associated with corresponding command functions according to actual needs. For example, after entering the voice command of "please shut down", the voice command is configured to be associated with the shutdown operation, so that the automatic shutdown of the biometric device under voice control can be realized. The embodiments of this specification do not specifically limit the categories and functions of voice commands, and can be set according to actual needs.

[0108] For example, voice commands can also include alarm setting commands, payment suspension commands, etc., among which, the alarm setting command can realize voice control of the biometric transaction device to set the alarm, while some work scenarios do not allow the use of mobile phones. By voice controlling the biometric device to set the alarm, the recipient can be reminded to handle related matters immediately, such as: the recipient can be reminded to ship or send express, or some shopping malls or supermarkets offer discounts that start at a specified time, and the alarm can remind the recipient to make timely adjustments to the payment operation. The payment suspension command can be used during the biometric recognition process. If the order is found to be wrong or the user does not want to buy, the voice-controlled biometric transaction device stops the biometric recognition and transaction payment process. If the payment has been completed, the payment suspension operation can realize voice control of the biometric device to refund the payment, so that the transaction can be suspended or the order can be canceled in time, which is simpler to operate.

[0109] The embodiments of this specification can associate voice commands with specific functional operations through voiceprint registration, and at the same time, store voiceprint features that can execute the voice commands, so that after receiving the voice command, identity recognition can be performed based on the voiceprint. After successful recognition, the operation corresponding to the voice command can be executed, which facilitates the control of the biometric transaction device and further facilitates the processing of transaction data.

[0110] In some embodiments of this specification, the method further includes:

[0111] receiving a voice query instruction, wherein the voice query instruction includes: a query time;

[0112] The voiceprint features in the voice query instruction are extracted, and the voiceprint features in the voice query instruction are matched with the target voiceprint features. If the match is successful, the transaction information of the biometric transaction device within the query time is queried according to the voice query instruction.

[0113] In the specific implementation process, the voice command can be a voice query command. After receiving the voice query command, the voiceprint features in the voice query command can be identified first, and the voiceprint features in the voice query command can be matched with the target voiceprint features. If the match is successful, based on the query time in the voice query command, the transaction information of the biometric transaction device within the query time is queried. Among them, the voice query command can save the corresponding voiceprint features and associate the voice query command with the query operation through the above-mentioned voiceprint registration process. Transaction information can be understood as transaction records or transaction amounts, etc.

[0114] In some embodiments of the present specification, the voice query instruction also includes the identity of the payment recipient to be queried, and the voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information of the biometric transaction device within the query time is queried according to the voice query instruction, and it also includes:

[0115] The voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information corresponding to the identity identifier within the query time in the biometric transaction device is queried according to the voice query instruction.

[0116] In the specific implementation process, the voice query instruction may include not only the query time, but also the identity of the payment object to be queried. The identity may be a unique identifier that is configured for the voiceprint feature when the voiceprint feature is registered and can identify the identity corresponding to the target voiceprint feature. Based on the identity in the voice query instruction, after the voiceprint feature of the voice query instruction successfully matches the target voiceprint feature, the transaction information corresponding to the identity in the biometric transaction device during the query time can be queried, thereby obtaining the transaction information of the payment object to be queried corresponding to the identity in the biometric device during the query time. Because each time the biometric transaction device starts the biometric function, it is controlled by the voice of the payment object, so each transaction data processed by the biometric transaction device corresponds to a payment object. Therefore, the voice query instruction can be used to query the transaction information of the specified payment object on the biometric transaction device based on the identity.

[0117] In addition, when using a voice query command to query the transaction information of a designated payment object, the identity identifier of the payment object that initiated the voice query command can also be matched with the identity identifier of the payment object to be queried. Only after the match is successful, the query is allowed, so that each payment object can only query its own transaction information. For example: after receiving the voice query command, the voiceprint feature in the voice query command can be obtained. After the voiceprint feature successfully matches the target voiceprint feature, the identity identifier corresponding to the target voiceprint feature matching the voiceprint feature is obtained. As described in the above embodiment, each target voiceprint feature can correspond to an identity identifier that can identify the identity of the payment object. The identity identifier corresponding to the target voiceprint feature is compared with the identity identifier of the payment object to be queried in the voice query command. If the two are consistent, the transaction information can be queried. If they are inconsistent, it means that the payment object corresponding to the target voiceprint feature is not the payment object to be queried, and it is not allowed to query the transaction information of the payment object to be queried, so as to ensure the security of the transaction information processed by each payment object.

[0118] Of course, the administrator of the payment device can generally allow the transaction information of all payment objects to be queried. The query permissions of each payment object can be set, such as: associating the identity of each payment object (that is, the identity corresponding to each target voiceprint feature) with the identity of the payment object to be queried that is allowed to be queried, so as to control the query operation of the voice query command and ensure the security of the transaction information.

[0119] The transaction data processing method provided in the embodiments of this specification can not only control the biometric transaction device by voice to start the biometric function to avoid transaction failure caused by the failure to complete the transaction order, but also can query the transaction information of the biometric transaction device within a specified time or the transaction information of the specified payment recipient on the biometric transaction device by voice, so as to verify the transaction information with the payment device to ensure the accuracy of the transaction.

[0120] In addition, the target voiceprint features corresponding to different voice commands can be stored in different voiceprint libraries respectively. When the biometric transaction device is controlled to perform the corresponding operation through voice commands, the content of the voice command can be extracted first, and the corresponding voiceprint library can be queried based on the voice command to match the voiceprint. For example, when registering the voiceprint, the voiceprint features corresponding to different voice commands can be saved respectively, so that when the biometric transaction device is controlled through voice commands, it can be more accurately identified whether the user who issued the command has the functional authority corresponding to the command. For example, cashier A can only voice control the biometric transaction device to turn on the biometric function. Then, when registering the voiceprint, the voiceprint features when cashier A initiates the voice start command can be saved in voiceprint library 1. Cashier B is the manager, who can not only control the voice control biometric transaction device to turn on the biometric function, but also query the transaction information of the biometric transaction device. Then, when registering the voiceprint, the voiceprint features of cashier B can be saved in voiceprint library 1 and voiceprint library 2. After receiving a voice command, the content of the voice command can be obtained first, and based on the content of the voice command, it can be determined whether to match the voiceprint features corresponding to the voice command with the target voiceprint features in voiceprint library 1, or with the target voiceprint features in voiceprint library 2. For example, if the content of the current voice command is recognized as a voice start command, the voiceprint features corresponding to the voice command can be matched with the target voiceprint features in voiceprint library 1, thereby improving the accuracy and security of voice command control of biometric transaction devices.

[0121] In some embodiments of this specification, a biometric transaction device for transaction data processing is also provided, the device comprising:

[0122] The voice collection module is used to collect voice start instructions; the voice start instructions are initiated by the payment recipient after the payment device constructs a transaction order, and the voice start instructions are used to start the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicates unidirectionally with the payment device;

[0123] Voiceprint command module, used for voiceprint registration, and the configuration of voice start command and command function association; voice start command is used to enable biometric recognition function;

[0124] A biometric feature collection module, used for collecting biometric features;

[0125] The transaction service module is used to perform voiceprint recognition on the voice start command collected by the voice collection device, and after the voiceprint feature matches successfully, based on the command function associated with the voice start command, turn on the biometric recognition function, and perform biometric recognition based on the biometrics collected by the biometrics collection module, obtain the payment account of the payment object corresponding to the biometrics, and complete the payment transaction based on the payment account and the virtual resources required for the transaction in the transaction order.

[0126] In the specific implementation process, Figure 3 is a schematic diagram of the structure of a biometric transaction device for transaction data processing in one embodiment of the present specification, such as Figure 3 As shown, the biometric transaction device may include a voice acquisition module, a voiceprint instruction module, a biometric feature acquisition module, a transaction service module, and a state control module. The following uses the biometric transaction device for palm-swipe payment transactions as an example to introduce the functions of the biometric transaction device:

[0127] 1) Voice collection module: In some embodiments of this specification, an array microphone can be installed on the biometric transaction device, and the array microphone can be used as a voice collection module to collect voice instructions issued by the cashier (i.e., the recipient of payment). The array microphone can collect voice in a directional manner. By utilizing the phase and amplitude differences between the sound waveforms received by each microphone in the multiple microphone arrays, the signal processing algorithm is used to analyze and calculate to determine the direction and position of the sound source, thereby achieving directional collection of the sound source and identifying the cashier's voice instructions. This technology can improve the signal-to-noise ratio and clarity of the voice signal and reduce the interference of environmental noise.

[0128] 2) Biometrics collection module: Figure 3 As shown, a camera (such as a 3D camera) can be provided on the biometric transaction device as a biometric feature collection module for collecting biometric features such as palm print features. The camera is mainly composed of a color sensor and an infrared sensor, that is, the color sensor of the RGB sensor and the IR sensor is used to take color images, and the infrared sensor is used to take infrared images. The algorithm can use dual factors for fusion recognition. At the same time, the infrared image taken by the infrared sensor can be used for biopsy, that is, to distinguish whether it is alive, so as to improve the accuracy of biometric feature collection.

[0129] In addition, if Figure 3As shown, a palm swiping module can be set in the biometric transaction device. The palm swiping module and the voiceprint command module and the control center described later can be integrated into an application such as: palm swiping APP (Application). The palm swiping module and the camera can be combined as a biometric feature acquisition module. For example, when collecting palm print features, the palm swiping APP can call the 3D camera to collect the user's current palm streaming media data. After obtaining the streaming media, the palm swiping module can optimize the streaming media. The optimization can be understood as selecting the best palm picture through a comprehensive evaluation of coefficient indicators such as palm size, angle, image contrast, image brightness and clarity. Subsequently, the palm swiping module sends the picture to the back-end service for palm recognition, and further obtains the user identity information or payment code related information corresponding to the palm.

[0130] 3) Voiceprint command module: The core includes two functions: voiceprint registration and voice command.

[0131] Voiceprint registration: When registering, users can record a sound sample containing multiple voice features, such as pitch, volume, speaking speed, pronunciation, etc. The sample length can be more than 10 seconds. These features will be extracted and used for the training of the voiceprint recognition model. During the recording and training process, attention should be paid to the consistency and stability of the environment to ensure the accuracy and reliability of the voiceprint recognition model. After the user successfully registers, the difference from palm swiping is that the sample recorded by the user can be stored on the current biometric transaction device instead of in the cloud. When the device is changed, the cashier needs to re-register. In this way, the voiceprint recognition can be completed locally on the biometric transaction device to confirm whether the identity of the user who initiated the voice command meets the requirements. The registered voiceprint does not need to be associated with the identity of other applications. When the cashier records the voiceprint registration, the terminal system can assign a unique identity in the current device dimension. The identity can be used to mark the identity of each registered user. The user can change it according to his or her needs, such as changing it to his or her own name, as long as it is not repeated. Figure 3 As shown, the registered voiceprint features can be stored in the local voiceprint library, or the voiceprint recognition model constructed based on the voiceprint feature training can be stored in the local voiceprint library to facilitate subsequent local voiceprint recognition.

[0132] Voice command: The system can require the cashier to provide a command voice sample and the command function corresponding to the voice command, such as: the voice start command corresponding to the start biometric recognition function, and the sample will also be used for voiceprint feature extraction. The extracted features will be compared with the voiceprint recognition model during the previous registration to determine the identity and authenticity of the user. If the voiceprint recognition is successful, the system will execute the corresponding command function. For example: to set the voice start command to "Please swipe your palm", the cashier needs to enter a voice sample of "Please swipe your palm" and configure the command function corresponding to the voice command to turn on the palm swiping function, that is, to let the palm swiping module enter the palm swiping state. Voice commands can be expanded according to the habits of cashiers. At the same time, voice commands can also be further expanded. For example, after entering the voice command of "Please shut down", the command can be configured to be associated with the shutdown operation.

[0133] 4) State control module: such as Figure 3 As shown, the biometric transaction device may also include a control center module, which is mainly used to control the device and perform operations according to the configuration. For example, voice commands can be configured through the control center, and the relationship between related commands and command functions can be stored, such as "please swipe your palm" and "the swiping device enters the swiping state". After turning on the swiping state, the front end can prompt the user to swipe their palm, and at the same time, the HID communication is turned on. After the palm print recognition is successful, it will directly spit out the code to the POS terminal. Among them, HID can be understood as a one-way communication mode. Generally, the palm swiping transaction device and the POS terminal are the one-way communication mode of HID, that is, the palm swiping device can send data to the POS terminal in one direction.

[0134] 5) Transaction service module: such as Figure 3 As shown, the transaction service module includes voiceprint recognition service, palmprint recognition service, and payment transaction service. The transaction service module can be integrated in the biometric transaction device, or it can be a server that can communicate data with the biometric transaction device, and this specification embodiment does not make specific limitations.

[0135] Voiceprint recognition service: Its core can be used to train voiceprint recognition models and convert the trained voiceprint recognition models into SDK (Software Development Kit) for biometric transaction devices. Subsequent voiceprint registration and voiceprint recognition only need to be performed on the palm-swipe transaction device (i.e., biometric transaction device), without going to the cloud. The voiceprint recognition model can be trained based on the Gaussian mixture model algorithm. The steps for model training can be referred to as follows:

[0136] a) Feature extraction: For the input speech signal, it must first be preprocessed, such as removing noise, reducing speech speed, etc. Then, the speech signal is divided into several frames, each frame usually contains 20-30 milliseconds of speech signal, and for each frame of speech signal, the Mel Frequency Cepstrum Coefficient (MFCC) and other algorithms can be used to extract its features to obtain a feature vector.

[0137] b) Establish a voiceprint recognition model

[0138] Multiple feature vectors are concatenated in chronological order to form a feature sequence. Then, the feature sequence is modeled using a Gaussian mixture model, which is a statistical model that can be used to model multidimensional data. In voiceprint recognition, Gaussian mixture models are usually used to model the probability distribution of sound features, that is, each feature vector is regarded as sampled from a Gaussian distribution. Usually, a voiceprint recognition model can include multiple Gaussian distributions, which is called a mixed Gaussian model. It can also be understood that a voiceprint recognition model can include multiple voiceprint recognition sub-models, each of which corresponds to a Gaussian distribution of voiceprint features. Figure 4 This is a data interaction diagram of the process of starting the palm payment function of the palm transaction device in a scenario example of this manual, such as Figure 4 As shown, after the cashier registers the voiceprint and enters the voiceprint and voice command, the recorded voiceprint and voice command can be sent to the back-end service (i.e., the transaction service module), and the voiceprint recognition model is trained and constructed through the back-end service.

[0139] c) Voiceprint recognition

[0140] For the input voice signal, feature extraction and preprocessing are also required. Then, the feature sequence is input into the previously established voiceprint recognition model, and the Bayesian theorem is used to calculate the posterior probability of the sequence corresponding to each voiceprint recognition sub-model, that is, the probability that the sequence comes from a certain voiceprint. Finally, the voiceprint with the largest posterior probability is selected as the recognition result.

[0141] Palm recognition service: Figure 4As shown, referring to the records of the above embodiments, after voiceprint registration and voice command and command function association are performed on the palm-swipe transaction device, the palm-swipe transaction device can continuously collect voice data in the environment. After the cashier creates a transaction order, the cashier can initiate a voice command. When the palm-swipe transaction device collects the voice command initiated by the cashier, the voiceprint features in the voice command can be extracted. After the above voiceprint recognition service is successfully matched, it is confirmed that the identity of the user who initiated the voice command meets the requirements, and then the palm-swipe function of the palm-swipe transaction device can be started, that is, the palm-swipe transaction device enters a state where it can identify the collected user's palm print features. After the palm-swipe transaction device collects the user's palm print features, the collected palm print features are identified, and the associated user's identity information such as payment account, etc. is obtained based on the user's palm print features.

[0142] Payment transaction service: can be used to complete payment transactions based on the user's identity information obtained through palm print recognition. Figure 5 This is a data interaction diagram of the entire palm payment process in a scenario example in this manual. Figure 5 As shown, based on the records of the above embodiments, after voiceprint registration and voice command association, the transaction order is created and completed on the payment device, and the cashier initiates a voice start command to enter a recognizable state. The user can perform palm swiping recognition, and the palm swiping transaction device can send the collected biometric features to the back-end service (i.e., the transaction service module). The back-end service obtains the identity information of the palm swiping user based on the collected palmprint information, such as the payment account. The payment transaction is then completed based on the payment account and the payment order created by the payment device, and the payment result is returned to the palm swiping transaction device. After receiving the payment result, the palm swiping transaction device can reset the status, such as turning off the palm swiping function, and returning the payment result to the user to prompt the user to view the transaction information.

[0143] In the embodiment of this specification, after the cashier creates a transaction order, the cashier's identity can be confirmed through voiceprint recognition, and the palm-swiping status can be transmitted to the palm-swiping device in a convenient way, so as to avoid the failure of the user's palm-swiping due to unreadiness, reduce the probability of repeated processing of transaction data in biometric payment transactions, and improve user experience and transaction data processing efficiency. In addition, by triggering the start of the biometric function through voice commands, there is no need for two-way communication between the cashier and the biometric transaction device, and there is no need for the cashier to cooperate in technical improvements, so that the biometric transaction can be completed accurately and quickly.

[0144] Based on the transaction data processing method described above, one or more embodiments of this specification also provide a terminal and a server for transaction data processing. The terminal and the server may include a device (including a distributed system), software (application), module, component, server, terminal, etc. using the method described in the embodiment of this specification and a device combined with necessary implementation hardware. Based on the same innovative concept, the device in one or more embodiments provided in the embodiment of this specification is as described in the following embodiments. Since the implementation scheme and method of the device to solve the problem are similar, the implementation of the specific device in the embodiment of this specification can refer to the implementation of the aforementioned method, and the repetitions will not be repeated. As used below, the term "unit" or "module" can implement a combination of software and / or hardware for predetermined functions. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0145] It can be seen from the technical solutions provided in the above embodiments of this specification that the embodiments of this specification also provide a transaction data processing device, Figure 6 is a schematic diagram of a transaction data processing device provided in an embodiment of the present specification, wherein the device is applied to a biometric transaction device, wherein the biometric transaction device is used to collect biometric features during the transaction data processing process, such as Figure 6 As shown, the above device comprises:

[0146] The voice start instruction receiving module 610 is used to receive a voice start instruction, which is initiated by the payment recipient after the payment device constructs a transaction order, and is used to start the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicates unidirectionally with the payment device;

[0147] The voiceprint recognition module 620 is used to extract the voiceprint features in the voice start instruction, match the voiceprint features in the voice start instruction with the stored target voiceprint features, and if the match is successful, start the biometric recognition function according to the voice start instruction;

[0148] A biometric feature recognition module 630, configured to collect the biometric features of the payment object using the biometric recognition function, and obtain the payment account of the payment object based on the collected biometric features;

[0149] The transaction processing module 640 is used to complete the payment transaction based on the payment account and the virtual resources required for the transaction in the transaction order, and to disable the biometric recognition function.

[0150] Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated in detail here. The device in the above embodiment may also include other implementation methods according to the description of the method embodiment, and the specific implementation method can refer to the description of the relevant method embodiment, which will not be repeated here one by one.

[0151] Figure 7 is a block diagram of an electronic device for transaction data processing provided by an embodiment of this specification. The electronic device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7 As shown. The electronic device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a transaction data processing method is implemented. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covering the display screen, or a key, trackball or touchpad set on the housing of the electronic device, or an external keyboard, touchpad or mouse, etc.

[0152] Figure 8 is a block diagram of another electronic device for transaction data processing provided by an embodiment of this specification. The electronic device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 As shown. The electronic device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a transaction data processing method is implemented.

[0153] Those skilled in the art will understand that Figure 7 or Figure 8 The structure shown in the figure is merely a block diagram of a portion of the structure related to the embodiment scheme of this specification, and does not constitute a limitation on the electronic device to which the embodiment scheme of this specification is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0154] In an exemplary embodiment, an electronic device is also provided, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the transaction data processing method as in the embodiment of this specification.

[0155] In an exemplary embodiment, a computer-readable storage medium is also provided. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the transaction data processing method in the embodiment of this specification.

[0156] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the transaction data processing method provided in the above various optional implementations.

[0157] It is understandable that in the specific implementation of this specification, user-related data is involved. When the above embodiments of this specification are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0158] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0159] Those skilled in the art will readily appreciate other implementations of the embodiments of the present specification after considering the specification and practicing the invention disclosed herein. The present specification is intended to cover any variations, uses, or adaptations of the embodiments of the present specification, which follow the general principles of the embodiments of the present specification and include common knowledge or customary technical means in the art that are not disclosed in the embodiments of the present specification. The description and embodiments are to be considered merely as exemplary, and the true scope and spirit of the embodiments of the present specification are indicated by the following claims.

[0160] It should be understood that the embodiments of the present specification are not limited to the precise structures described above and shown in the drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the embodiments of the present specification is limited only by the appended claims.

Claims

1. A transaction data processing method, characterized in that: The method is applied to a biometric transaction device, which is used to collect biometric features during transaction data processing. The method includes: receiving a voice start instruction, the voice start instruction being initiated by the payment recipient after the payment device constructs a transaction order, the voice start instruction being used to start the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicating unidirectionally with the payment device; Extracting the voiceprint feature in the voice start instruction, matching the voiceprint feature in the voice start instruction with the stored target voiceprint feature, and if the match is successful, starting the biometric recognition function according to the voice start instruction; Using the biometric recognition function to collect biometric features of the payment object, and obtaining the payment account of the payment object based on the collected biometric features; Based on the payment account and the virtual resources required for the transaction in the transaction order, the payment transaction is completed and the biometric recognition function is turned off.

2. The method according to claim 1, characterized in that The receiving of the voice start instruction comprises: Using an array microphone provided in the biometric transaction device to collect voice data, and determining the sound source position and direction corresponding to the collected voice data according to the array microphone; Based on the sound source position and direction corresponding to the voice data, determining the target voice data belonging to the location of the payment object from the voice data; The target voice data is used as the voice start instruction.

3. The method according to claim 1, characterized in that The extracting the voiceprint feature in the voice start instruction, matching the voiceprint feature in the voice start instruction with the stored target voiceprint feature, and if the match is successful, starting the biometric recognition function according to the voice start instruction, includes: Extracting a voiceprint feature vector corresponding to each frame of the voice signal in the voice start instruction; The extracted voiceprint feature vectors are concatenated in chronological order to obtain a voiceprint feature sequence; The voiceprint feature sequence is input into a pre-trained voiceprint recognition model, and the voiceprint recognition model is used to calculate the matching probability between the voiceprint feature sequence and each of the target voiceprint features. If the matching probability between the voiceprint feature sequence and at least one target voiceprint feature is greater than a preset threshold, the match is determined to be successful, and the biometric recognition function is started according to the voice start instruction; wherein the voiceprint recognition model is constructed based on the training of the target voiceprint features.

4. The method according to claim 3, characterized in that: The method for constructing the voiceprint recognition model includes: Extracting a sample voiceprint feature vector from each frame of voice signal in each voice sample data; wherein each voice sample data is voice data of a target object that is allowed to activate the biometric recognition function of the biometric recognition transaction device; The sample voiceprint feature vectors corresponding to each voice sample data are concatenated in chronological order to obtain a sample voiceprint feature sequence corresponding to each voice sample data, and the sample voiceprint feature sequence is used as the target voiceprint feature; The probability distribution of the target voiceprint feature is extracted using a Gaussian mixture model, and the voiceprint recognition model is trained and constructed; wherein the voiceprint recognition model includes multiple voiceprint recognition sub-models, and each voiceprint recognition sub-model corresponds to a Gaussian distribution of the target voiceprint feature.

5. The method according to claim 1, characterized in that Before receiving the voice start instruction, the method further includes: Receiving a voiceprint registration request initiated by the payment recipient, wherein the voiceprint registration request includes voice sample data of the payment recipient, a voice start instruction, and an instruction function corresponding to the voice start instruction, wherein the instruction function is to start the biometric recognition function of the biometric recognition transaction device; Extracting target voiceprint features from the voice sample data and storing the target voiceprint features; The voice activation instruction is associated with an instruction function for activating a biometric function of the biometric transaction device.

6. The method according to claim 1, characterized in that The method further comprises: receiving a voice query instruction, wherein the voice query instruction includes a query time; The voiceprint feature in the voice query instruction is extracted, and the voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information of the biometric transaction device within the query time is queried according to the voice query instruction.

7. The method according to claim 6, characterized in that The voice query instruction also includes the identity of the payment recipient to be queried, and the voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information of the biometric transaction device within the query time is queried according to the voice query instruction, and further includes: The voiceprint feature in the voice query instruction is matched with the target voiceprint feature. If the match is successful, the transaction information corresponding to the identity identifier within the query time in the biometric transaction device is queried according to the voice query instruction.

8. A transaction data processing device, characterized in that: The device is applied to a biometric transaction device, and the biometric transaction device is used to collect biometric features during transaction data processing. The device includes: A voice start instruction receiving module, used for receiving a voice start instruction, wherein the voice start instruction is initiated by the payment recipient after the payment device constructs a transaction order, and the voice start instruction is used for starting the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicates unidirectionally with the payment device; A voiceprint recognition module, used to extract the voiceprint features in the voice start instruction, match the voiceprint features in the voice start instruction with the stored target voiceprint features, and if the match is successful, start the biometric recognition function according to the voice start instruction; A biometric feature recognition module, used to collect the biometric features of the payment object using the biometric recognition function, and obtain the payment account of the payment object based on the collected biometric features; The transaction processing module is used to complete the payment transaction based on the payment account and the virtual resources required for the transaction in the transaction order, and to close the biometric recognition function.

9. A biometric transaction device for transaction data processing, characterized in that: The device comprises: A voice collection module, used for collecting voice start instructions; the voice start instructions are initiated by the payment recipient after the payment device constructs a transaction order, and the voice start instructions are used to start the biometric recognition function of the biometric recognition transaction device; the biometric recognition transaction device communicates unidirectionally with the payment device; Voiceprint command module, used for voiceprint registration and the configuration of the association between voice start commands and command functions; A biometric feature collection module, used for collecting biometric features; a transaction service module, configured to perform voiceprint recognition on the voice start instruction collected by the voice collection device, match the voiceprint feature in the voice start instruction with the target voiceprint feature, and after the biometric recognition function is turned on, perform biometric recognition based on the biometric feature collected by the biometric feature collection module, obtain the payment account of the payment object corresponding to the biometric feature, and complete the payment transaction based on the payment account and the virtual resources required for the transaction in the transaction order; The state control module is used to enable the biometric recognition function based on the instruction function associated with the voice start instruction after the transaction service module determines that the voiceprint feature in the voice start instruction is successfully matched.

10. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the transaction data processing method according to any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the transaction data processing method as described in any one of claims 1 to 7.

Images