Data processing method, device, electronic device and storage medium

Abstract

This application discloses a data processing method, apparatus, electronic device, and storage medium. The method includes: acquiring target biometric information when a primary identity identifier is valid; identifying a business intent and a target user identity corresponding to the target biometric information; acquiring a target secondary identity identifier corresponding to the target user identity; the target secondary identity identifier being a sub-identifier of the primary identity identifier; and executing a business instruction corresponding to the business intent based on the target secondary identity identifier. Using this application, the business behavior executed by the terminal device can be matched with the user identity.

Description

Technical Field

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

[0002] With the rapid development of Internet technology, the number of Internet users is also growing. Among all application software, video software is one of the most frequently used software by Internet users. Data shows that the usage time of video software accounts for as much as 34.5% of the total usage time of mobile devices.

[0003] In a family scenario, terminal devices (for example, smart TVs) are often shared by multiple family members. When family member A uses his account A to log in to the video application in the terminal device and watches video 1, the terminal device will record the viewing progress of family member A on video 1. If another family member B uses the video application to watch video 1 again and does not switch the login account of the video application, the terminal device automatically jumps to the viewing progress of family member A on video 1, causing the business behavior executed by the terminal device to not match the user identity. Summary of the invention

[0004] The embodiments of the present application provide a data processing method, apparatus, and related equipment, which can match the business behavior executed by a terminal device with the user identity.

[0005] On the one hand, an embodiment of the present application provides a data processing method, including:

[0006] When the primary identity is in a valid state, the target biological information is obtained;

[0007] Identifying the business intent and target user identity corresponding to the target biometric information;

[0008] Acquire a target secondary identity identifier corresponding to the target user identity; the target secondary identity identifier is a sub-identifier of the primary identity identifier;

[0009] Execute a business instruction corresponding to the business intention based on the target secondary identity.

[0010] Wherein, the target biological information includes target voice data;

[0011] The identifying the business intention and target user identity corresponding to the target biometric information includes:

[0012] Convert the target voice data into text data, semantically recognize the text data, and obtain the business intent;

[0013] Calling an identity recognition model corresponding to the primary identity identifier to determine a matching result between the target voice data and at least one template user identity; the identity recognition model is a classification model generated based on the at least one template user identity and the template voice data corresponding to the at least one template user identity;

[0014] If there is a matching result satisfying the matching condition in at least one matching result, the template user identity corresponding to the matching result satisfying the matching condition is used as the target user identity;

[0015] The step of obtaining a target secondary identity corresponding to the target user identity includes:

[0016] A target identity corresponding to the target user identity is extracted from a set of secondary identity identifiers corresponding to the at least one template user identity; the secondary identity identifiers in the set of secondary identity identifiers are sub-identifiers of the primary identity identifier.

[0017] Among them, it also includes:

[0018] If there is no matching result satisfying the matching condition in the at least one matching result, creating the target user identity;

[0019] Identify age information corresponding to the target voice data, and search for an identity head image matching the age information in an image material library;

[0020] The step of obtaining a target secondary identity corresponding to the target user identity includes:

[0021] Creating the target secondary user identifier for the target user identity;

[0022] Setting the target secondary user identifier as a sub-identifier of the primary identity identifier;

[0023] The target user identity, the target secondary identity identifier and the identity avatar are stored in association.

[0024] Wherein, the identity recognition model includes a feature generator and a pattern matcher;

[0025] The calling of the identity recognition model corresponding to the primary identity identifier to determine a matching result between the target voice data and at least one template user identity includes:

[0026] Based on the feature generator, extracting target voiceprint features of the target voice data;

[0027] Based on the pattern matcher, the matching probability between the target voiceprint feature and at least one template voiceprint feature is determined, and the obtained matching probabilities are all used as matching results; the at least one template voiceprint feature is the voiceprint feature corresponding to the at least one template voice data respectively.

[0028] The step of extracting the target voiceprint features of the target voice data based on the feature generator includes:

[0029] Based on the feature generator, extract the spectrum parameters and linear prediction parameters of the target speech data; the spectrum parameters are the short-time spectrum feature parameters of the target speech data; the linear prediction parameters are the spectrum fitting feature parameters of the target speech data;

[0030] The target voiceprint feature is obtained according to the spectrum parameter and the linear prediction parameter.

[0031] Among them, it also includes:

[0032] Obtain template voice data corresponding to the template user identity;

[0033] Generating an identity tag vector corresponding to the template speech data;

[0034] Acquire an initial classification model, predict a matching degree between the sample voice data and the at least one template user identity based on the initial classification model, and obtain an identity prediction vector according to the acquired matching degree;

[0035] A classification error is determined according to the identity label vector and the identity prediction vector, and the initial classification model is trained according to the classification error to obtain the identity recognition model.

[0036] Among them, it also includes:

[0037] When there is a matching result satisfying the matching condition in the at least one matching result, sending an animation playing instruction to the client, instructing the client to play the target animation;

[0038] When the execution of the service instruction is completed, a stop animation playing instruction is sent to the client, instructing the client to close the target animation.

[0039] Wherein, the business intention includes the client secondary login object switching intention;

[0040] The executing the business instruction corresponding to the business intention based on the target secondary identity includes:

[0041] Generate a switching instruction corresponding to the secondary login object switching intention of the client; the switching instruction belongs to the business instruction;

[0042] According to the switching instruction, the target secondary identity is used as a secondary login object of the client.

[0043] Among them, it also includes:

[0044] Acquire behavior data of a user corresponding to the target secondary identity in the client; the behavior data is used to generate recommended service data for the user;

[0045] The behavior data and the target secondary identity are associated and stored.

[0046] Wherein, the business intention includes business data query intention;

[0047] The executing the business instruction corresponding to the business intention based on the target secondary identity includes:

[0048] Generate a query instruction corresponding to the business data query intention; the query instruction belongs to the business instruction;

[0049] Query the target business data corresponding to the target secondary identity identifier, and return the target business data to the client.

[0050] The user rights of the target secondary identity are the same as the user rights of the primary identity.

[0051] Another aspect of the present application provides a data processing device, including:

[0052] A first acquisition module, used to acquire target biological information when the primary identity identifier is in a valid state;

[0053] An identification module, used to identify the business intention and target user identity corresponding to the target biometric information;

[0054] A second acquisition module is used to acquire a target secondary identity identifier corresponding to the target user identity; the target secondary identity identifier is a sub-identifier of the primary identity identifier;

[0055] A determination module is used to execute a business instruction corresponding to the business intention based on the target secondary identity.

[0056] Wherein, the target biological information includes target voice data;

[0057] The identification module comprises:

[0058] A conversion unit, used to convert the target voice data into text data, semantically identify the text data, and obtain the business intention;

[0059] A calling unit, configured to call an identity recognition model corresponding to the primary identity identifier to determine a matching result between the target voice data and at least one template user identity; the identity recognition model is a classification model generated based on the at least one template user identity and the template voice data corresponding to the at least one template user identity;

[0060] a first determining unit, configured to use, if there is a matching result satisfying a matching condition in at least one matching result, a template user identity corresponding to the matching result satisfying the matching condition as the target user identity;

[0061] Then the second acquisition module includes:

[0062] The first extraction unit is used to extract a target identity corresponding to the target user identity from a set of secondary identity corresponding to the at least one template user identity; the secondary identity in the set of secondary identity is a sub-identifier of the primary identity.

[0063] Among them, it also includes:

[0064] a second determining unit, configured to create the target user identity, identify age information corresponding to the target voice data, and search an identity avatar matching the age information in an image material library if no matching result that satisfies the matching condition exists in the at least one matching result;

[0065] Then the second acquisition module includes:

[0066] The second extraction unit is used to create the target secondary user identifier for the target user identity, set the target secondary user identifier as a sub-identifier of the primary identity identifier, and associate and store the target user identity, the target secondary identity identifier, and the identity avatar.

[0067] Wherein, the identity recognition model includes a feature generator and a pattern matcher;

[0068] The calling unit comprises:

[0069] An extraction subunit, configured to extract a target voiceprint feature of the target speech data based on the feature generator;

[0070] The matching subunit is used to determine the matching probability between the target voiceprint feature and at least one template voiceprint feature based on the pattern matcher, and use the obtained matching probabilities as matching results; the at least one template voiceprint feature is the voiceprint feature corresponding to the at least one template voice data respectively.

[0071] Among them, the extraction subunit is specifically used to extract the spectral parameters and linear prediction parameters of the target speech data based on the feature generator, and obtain the target voiceprint features according to the spectral parameters and the linear prediction parameters; the spectral parameters are the short-time spectrum feature parameters of the target speech data; the linear prediction parameters are the spectrum fitting feature parameters of the target speech data.

[0072] Among them, it also includes:

[0073] A training module is used to obtain template voice data corresponding to a template user identity, generate an identity label vector corresponding to the template voice data, obtain an initial classification model, predict the degree of match between the template voice data and at least one template user identity based on the initial classification model, obtain an identity prediction vector according to the obtained degree of match, determine a classification error according to the identity label vector and the identity prediction vector, train the initial classification model according to the classification error, and obtain the identity recognition model.

[0074] Among them, it also includes:

[0075] A playing module, configured to send an animation playing instruction to a client when there is a matching result satisfying the matching condition in the at least one matching result, so as to instruct the client to play a target animation;

[0076] The playing module is further used to send a stop animation playing instruction to the client when the business instruction is executed, instructing the client to close the target animation.

[0077] Wherein, the business intention includes the client secondary login object switching intention;

[0078] The determining module comprises:

[0079] The first generating unit is used to generate a switching instruction corresponding to the secondary login object switching intention of the client, and according to the switching instruction, the target secondary identity identifier is used as the secondary login object of the client; the switching instruction belongs to the business instruction.

[0080] Among them, it also includes:

[0081] The storage module is used to obtain the behavior data of the user corresponding to the target secondary identity in the client, and store the behavior data and the target secondary identity in association; the behavior data is used to generate recommended business data for the user.

[0082] Wherein, the business intention includes business data query intention;

[0083] The determining module comprises:

[0084] The second generating unit is used to generate a query instruction corresponding to the business data query intention, query the target business data corresponding to the target secondary identity identifier, and return the target business data to the client; the query instruction belongs to the business instruction.

[0085] The user rights of the target secondary identity are the same as the user rights of the primary identity.

[0086] Another aspect of an embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes a method as described in one aspect of an embodiment of the present application.

[0087] On the other hand, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the method in one aspect of the embodiment of the present application is executed.

[0088] The present application can determine the target secondary identity corresponding to the user identity by identifying the user identity and business intention that generates the current biometric information. Therefore, the business instructions executed based on the target secondary identity not only meet the user's current business intention, but also match the user identity. Furthermore, the present application only needs to collect the user's target biometric information, and can simultaneously determine the user's business intention and user identity. The user does not have to perform two operations, namely, determining the business intention and determining the user identity, which can reduce the user's operating costs and improve the efficiency of the terminal in executing business instructions that match both the user's business intention and user identity. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0090] Figure 1 It is a system architecture diagram of data processing provided by an embodiment of the present application;

[0091] Figure 2a-2d It is a schematic diagram of a data processing scenario provided by an embodiment of the present application;

[0092] Figure 3 It is a flowchart of a data processing method provided in an embodiment of the present application;

[0093] Figure 4 It is a flowchart of another data processing method provided in an embodiment of the present application;

[0094] Figure 5 is a timing diagram of a data processing method provided in an embodiment of the present application;

[0095] Figure 6 It is a flow chart of determining a target user identity and a target secondary identity provided by an embodiment of the present application;

[0096] Figure 7 It is a flowchart of another data processing method provided in an embodiment of the present application;

[0097] Figure 8 is a timing diagram of another data processing method provided in an embodiment of the present application;

[0098] Fig. 9 is a structural schematic diagram of a data processing device provided in an embodiment of the present application;

[0099] Fig.10 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0100] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0101] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0102] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. The basic technologies of artificial intelligence generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0103] The solution provided in the embodiment of the present application belongs to speech technology (Speech Technology), natural language processing (Nature Language processing, NLP) and machine learning (Machine Learning, ML) under the field of artificial intelligence.

[0104] The key technologies of speech technology include automatic speech recognition technology (ASR), text-to-speech technology (TTS) and voiceprint recognition technology, which enable computers to listen, see, speak and feel.

[0105] Natural language processing (NLP) is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that can achieve effective communication between people and computers using natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field will involve natural language, that is, the language people use in daily life, so it is closely related to the study of linguistics. Natural language processing technology usually includes text processing, semantic understanding, machine translation, robot question answering, knowledge graph and other technologies.

[0106] In this application, voice technology is involved in converting the user's voice into text, and natural language processing is involved in semantically recognizing text to determine the user's intention.

[0107] Machine learning is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithmic complexity theory and other disciplines. It specializes in studying how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. In this application, machine learning is related to identifying the user identity of the current user, and the specific technical means involve artificial neural networks in machine learning and logistic regression and other technologies.

[0108] See also Figure 1, is a system architecture diagram of a data processing provided by an embodiment of the present application. The present application involves a server 10d and a terminal device cluster, and the terminal device cluster may include: a terminal device 10a, a terminal device 10b, ..., a terminal device 10c, etc.

[0109] Taking the terminal device 10a as an example, when the primary identity is in a valid state, the terminal device 10a collects the user's biometric information and sends the collected biometric information to the server 10d. The server 10d performs semantic recognition on the biometric information to determine the intent of the biometric information; the server 10d also determines the user identity of the biometric information and extracts the secondary identity of the user identity. The extracted secondary identity is a sub-identifier of the primary identity. The server 10d executes instructions related to the above intent based on the determined secondary identity. Subsequently, the server 10d can return the execution result of the instruction to the terminal device 10a.

[0110] Identifying the intent of the biometric information, determining the identity of the user of the biometric information, and executing instructions related to the intent may also be accomplished by the terminal device 10a.

[0111] in, Figure 1 The terminal devices 10a, 10b, ..., 10c, etc. shown may include smart TVs, mobile phones, tablet computers, laptop computers, PDAs, mobile Internet devices (MIDs) or wearable devices (such as smart watches, smart bracelets, etc.), etc. Figure 1 The server 10d shown may refer to a single server device or a server cluster including a plurality of server devices.

[0112] The following Figure 2a-2d How the terminal device 10a identifies the intention of biometric information, determines the user identity of the biometric information, and executes instructions related to the intention is used as an example for specific description, and the identification of the intention of biometric information, the determination of the user identity of the biometric information, and the execution of instructions related to the intention can be specifically a video client in the terminal device 10a:

[0113] See also Figure 2a-2d , is a schematic diagram of a data processing scenario provided by an embodiment of the present application. The current user starts the video client in the terminal device 10a. When the video client detects that the first-level account "01" has been logged in, but there is no second-level account under the first-level account "01" logged in, a prompt message can be displayed on the screen: "The second-level account is not currently logged in. You can log in by voice input or clicking here", to prompt the current user to log in to the second-level account, where the first-level account "01" is the first-level account of user 1.

[0114] The current user can input by voice: "Log in to the secondary account", and the video client can obtain the voice data 20b of the voice "Log in to the secondary account". The video client converts the voice data 20b into text data, semantically recognizes the text data, and determines that the intention corresponding to the voice data 20b is: "Secondary account login".

[0115] The video client inputs the voice data 20b into the prediction model 20d that has been trained corresponding to the first-level account "01". The prediction model 20d can extract the voiceprint features of the voice data 20b, and match the voiceprint features with multiple template voiceprint features. If there is a template voiceprint feature among the multiple template voiceprint features that matches the voiceprint features of the voice data 20b, the user identity corresponding to the matched template voiceprint feature is extracted (assuming that the extracted user identity is user 2).

[0116] Each template voiceprint feature in the prediction model 20d corresponds to one user identity. Assume that the prediction model 20d is trained by two template voiceprint features. The two template voiceprint features correspond to user identities: user 1 and user 2, respectively. The secondary account of user 1 and the secondary account of user 2 are both sub-accounts of the current primary account "01".

[0117] like Figure 2b As shown, the secondary account number corresponding to user 2 is searched in the user information record table 20e corresponding to the primary account number "01" to find: 002.

[0118] from Figure 2b It can be seen that the user information record table 20e includes 3 user records, which correspond to a first-level account "01" and two second-level accounts under the first-level account "01" (respectively, the second-level account "001" and the second-level account "002"); the first-level account "01" and the second-level account "001" are both the accounts of user 1, and the second-level account "002" is the account of user 2; the user's historical records are all stored in association with the second-level accounts.

[0119] The video client extracts the secondary account of user 2: 002. Since the intention of the voice data 20b determined above is "secondary account login", the secondary account "002" can be used as the secondary login object of the video client.

[0120] like Figure 2b As shown in page 20f, the video client can display an animation on the screen during the process of determining the intention of the voice data 20b and determining the secondary account "002". When the video client has used the secondary account "002" as the secondary login object, the animation is stopped and the video client is redirected to its homepage.

[0121] As shown in page 20g, the video client is currently logged in to the secondary account "002" and the primary account "01".

[0122] Optionally, the above assumes that there is a template voiceprint feature among the multiple template voiceprint features that matches the voiceprint feature of the voice data 20b, thereby determining that the user identity corresponding to the voice data 20b is user 2.

[0123] like Figure 2c As shown, it is assumed that there is no template voiceprint feature matching the voiceprint feature of voice data 20b among the multiple template voiceprint features, that is, the current user who generates voice data 20b does not have a corresponding user identity, and there is no corresponding user record in the user information record table 20c. Since there is no corresponding user record, the video client can create a new user record for the current user, and the user record includes: user identity "User 2", secondary account "002", avatar, level "Level 2" and historical records (of course, the historical records are empty at this time).

[0124] The video client adds the user record to the user information record table 20c, and after the addition, a new user information record table 20h can be obtained.

[0125] The video client extracts the newly created secondary account of user 2: 002. Since the intention of the voice data 20b determined above is "secondary account login", the secondary account "002" can be used as the secondary login object of the video client.

[0126] like Figure 2d As shown in page 20i, after the video client creates a secondary account "002" and logs in to the secondary account "002", a prompt message can be displayed on the screen: "Your secondary account was not detected, a new secondary account was created for you, and you have logged in", to prompt the current user that a new secondary account has been created.

[0127] As shown in page 20j, the video client is currently logged in to the newly created secondary account "002" and the primary account "01".

[0128] The specific process of obtaining target biometric information (such as the voice data 20b of the voice "login to a secondary account" in the above embodiment), identifying the service intention (such as the intention "login to a secondary account" in the above embodiment) and the target user identity (such as user 2 in the above embodiment) can be referred to as follows Figure 3-Figure 8 The corresponding embodiments.

[0129] See also Figure 3 , is a flow chart of a data processing method provided in an embodiment of the present application, such as Figure 3As shown, the data processing method may include the following steps:

[0130] Step S101, when the primary identity is in a valid state, obtaining target biological information.

[0131] Specifically, the server (as mentioned above Figure 1 The server 10d in the corresponding embodiment detects whether the primary identity identifier is the current corresponding client (as described above). Figure 2a-2d The first-level login object of the video client in the corresponding embodiment, if the first-level identity is the first-level login object of the client, it means that the first-level identity is in a valid state. The login object of the client can include a first-level login object and a second-level login object. The first-level login object corresponds to the first-level identity, and the second-level login object corresponds to the second-level identity. The second-level identity is a sub-identity of the first-level identity.

[0132] The client may specifically be a video client, an instant messaging client, or an email client, etc.

[0133] When the primary identity is in a valid state, the server can receive the biometric information sent by the client (called the target biometric information, as mentioned above). Figure 2a-2d The voice data 20b) of the voice corresponding to the embodiment "Log in to the secondary account".

[0134] The target biometric information may include voice data (referred to as target voice data), and the target biometric information may also include voice data (referred to as target voice data) and image data (referred to as target image data), wherein the target image data may be facial image data of the current user.

[0135] Step S102: identifying the business intention and target user identity corresponding to the target biometric information.

[0136] Specifically, when the target biometric information includes target voice data, the server may convert the target voice data into text data and semantically identify the text data to determine the business intent of the target voice data.

[0137] And, the server can use the identity recognition model corresponding to the first identity identifier (such as the above Figure 2a-2d The prediction model 20d in the corresponding embodiment determines the user identity of the current user (referred to as the target user identity, as described above). Figure 2a-2d Corresponding to user 2) in the embodiment.

[0138] There is no restriction on the order in which the server determines the service intent and the target user identity.

[0139] Among them, the target speech data can be converted into text data by using an acoustic model (the acoustic model can be a model established based on a dynamic time warping method based on pattern matching, or a model established based on an artificial neural network recognition method, etc.), determining the state of each audio frame of the target speech data, merging several states into phonemes, and then merging several phonemes into words.

[0140] A language model (the language model may be an N-Gram language model, a Markov N-gram model, an exponential model, or a decision tree model, etc.) is used to combine multiple words into correct, unambiguous, and logical sentences to obtain text data.

[0141] Semantically identify the text data to determine the business intent of the text data. The entity-predicate knowledge graph can be used to perform pattern matching with the text data to determine the entities and predicates in the text data. The server can combine the identified entities and predicates into business intent.

[0142] For example, the current user voice input is: "Query the historical playback records". After converting the above voice data into text data "Query the historical playback records", the knowledge graph can be used to determine the entity as: "Historical playback records", the predicate is "query", and therefore the business intent is: historical playback records-query.

[0143] The identity recognition model is a classification model trained by at least one template user identity and the voice data corresponding to the template user identity (referred to as template voice data). Each template user identity has a corresponding secondary identity identifier (such as the above Figure 2a-2d Corresponding to the secondary account "001" and the secondary account "002" in the embodiment, the identity identifier may be a user account), and the secondary identity identifier of each template user identity is a sub-identifier of the primary identity identifier.

[0144] When the target biometric information includes target voice data and target image data, the server may also determine the business intent of the target voice data in the above manner, and determine the user identity of the target voice data (referred to as the first user identity) according to the identity recognition model;

[0145] The server can also determine the user identity of the target image data (referred to as the second user identity) based on the image recognition model. The image recognition model is similar to the identity recognition model in that both are classification models trained by at least one template user identity and the image data corresponding to the template user identity (referred to as template image data).

[0146] The server may determine the final target user identity based on the first user identity determined by the identity recognition model and the second user identity determined by the image recognition model. The target user identity determined based on the two models has a higher accuracy rate.

[0147] Alternatively, when the target biometric information includes target voice data and target image data, the server can also determine the business intent of the target voice data in the above manner, and determine the target user identity of the target image data based only on the image recognition model.

[0148] Step S103, obtaining a target secondary identity identifier corresponding to the target user identity; the target secondary identity identifier is a sub-identifier of the primary identity identifier.

[0149] Specifically, the server may obtain a target secondary identity identifier of the target user identity from a set of secondary identity identifiers corresponding to at least one template user identity (such as the above Figure 2a-2d The secondary account "002" in the corresponding embodiment) or a target secondary identity is recreated, and the target secondary identity is a sub-identity of the primary identity.

[0150] Step S104: executing a business instruction corresponding to the business intention based on the target secondary identity.

[0151] Specifically, when the business intention is the client's secondary login object switching intention, the server generates a switching instruction corresponding to the client's secondary login object switching intention. The switching instruction is used to instruct the server to switch the client's current secondary login object; the switching instruction belongs to a business instruction.

[0152] The server can use the target secondary identity as the current secondary login object of the client according to the switching instruction. Subsequently, the server can send a switching notification message to the client, so that after receiving the switching notification message, the client can display a prompt message to prompt the user that the current secondary login object is the target secondary identity.

[0153] Subsequently, when the client's secondary login object is the target secondary identity, the server can receive behavior data reported by the client (the behavior data may include: viewing behavior data, browsing behavior data, search behavior data, and at least one of comment behavior data), where the behavior data is user behavior data of the user collected by the client when the client's secondary login object is the target secondary identity.

[0154] The server may associate and store the target secondary identity with the behavior data reported by the client. Subsequently, the server may generate recommended service data for the user based on the behavior data to achieve the purpose of personalized recommendation.

[0155] When the business intent is a business data query intent, the server generates a query instruction corresponding to the business data query intent, and the query instruction is used to instruct the server to query business data; the query instruction belongs to a business instruction. For example, querying historical viewing records, querying viewing progress records, querying search records, etc.

[0156] The server can query the business data related to the target secondary identity (referred to as target business data) according to the query instruction. Subsequently, the server can return the queried target business data to the client, so that the client can display the target business data after receiving the target business data.

[0157] Alternatively, after the server generates the query instruction, it may first use the target secondary identity as the secondary login object of the client, and at the same time, execute the query operation related to the query instruction.

[0158] It should be noted that the user rights of the target secondary identity are the same as those of the primary identity. Furthermore, the user rights of all sub-identities of the primary identity are the same as those of the primary identity.

[0159] For example, if the primary identity has member VIP privileges, then the sub-identities of the primary identity (including the target secondary identity and the secondary identity of the template user identity mentioned above) all have member VIP privileges.

[0160] Different levels of user identification correspond to different functional architectures. The first-level identification can be used to manage member permissions and to collect statistical information of all second-level identifications (for example, total viewing time, etc.). The second-level identification is used to manage personalized information for each user identity.

[0161] It should be noted that the above steps S101 to S104 are described with the server as the execution subject, and the execution subject may also be installed on the terminal device (such as the above Figure 2a-2d The client in the terminal device 10a) in the corresponding embodiment may be a smart TV, and the client may be a video client installed in the smart TV.

[0162] When the primary identity is in a valid state, the client collects the target biometric information, and identifies the business intent of the target biometric information and calls the identity recognition model to determine the target user identity; the client obtains the target secondary identity of the target user identity, and executes business instructions corresponding to the business intent based on the target secondary identity, for example, a business instruction to use the target secondary identity as the secondary login object of the client, a business instruction to query the target business data corresponding to the target secondary identity, etc.

[0163] The present application can determine the target secondary identity corresponding to the user identity by identifying the user identity and business intention that generates the current biometric information. Therefore, the business instructions executed based on the target secondary identity not only meet the user's current business intention, but also match the user identity. Furthermore, the present application only needs to collect the user's target biometric information, and can simultaneously determine the user's business intention and user identity. The user does not have to perform two operations, namely, determining the business intention and determining the user identity, which can reduce the user's operating costs and improve the efficiency of the terminal in executing business instructions that match both the user's business intention and user identity.

[0164] See also Figure 4 , is a flow chart of another data processing method provided in an embodiment of the present application, wherein the data processing comprises the following steps:

[0165] Step S201, the process starts.

[0166] Step S202: The server obtains voice data.

[0167] Specifically, the current first-level account (which can correspond to the first-level identity identifier in this application) has logged in to the client, that is, the first-level account is in a valid state, and the server receives voice data sent by the client, where the voice data is the data collected by the client when the user voice inputs "login sub-account" in the client.

[0168] Step S203: the server determines whether the voiceprint already exists.

[0169] Specifically, the server semantically recognizes the voice data and determines that the business intention is to log in to the sub-account.

[0170] The server calls the identity recognition model of the first-level account to determine whether there is a template voiceprint feature that matches the voiceprint feature of the above-mentioned voice data. If so, execute steps S204 and S206; if not, execute steps S205-S206.

[0171] Step S204: the server logs the secondary account into the client.

[0172] Specifically, the server sets the secondary account corresponding to the matched voiceprint feature (which may correspond to the target secondary identity in this application) as the secondary login account of the client. At this time, one primary account and one secondary account are logged in to the client.

[0173] In step S205, the server creates a new secondary account (which may correspond to the target secondary identity in this application), and the secondary account is a sub-account of the primary account, associates and stores the newly created secondary account with the voiceprint feature, and uses the newly created secondary account as the secondary login account of the client.

[0174] Step S206, end the process.

[0175] Please see further Figure 5 , is a timing diagram of a data processing method provided in an embodiment of the present application. The video background server, speech recognition server and voiceprint recognition server described below all belong to the server in the present application. The data processing includes the following steps:

[0176] Step S301, the primary account is in a valid state, and the client collects the voice data input by the user to "enter the sub-account".

[0177] Step S302: The client sends the voice data to the video background server.

[0178] Step S303: the video background server sends the voice data to the voice recognition server.

[0179] Step S304: the video background server sends the voice data to the voiceprint recognition server.

[0180] Step S305 , the speech recognition server performs semantic recognition on the speech data, determines that the business intention is to access the secondary account, and sends the determined business intention back to the video background server.

[0181] Step S306: The voiceprint recognition server performs voiceprint recognition on the voice data according to the identity recognition model corresponding to the primary account to obtain a voiceprint recognition result, and the voiceprint recognition server sends the voiceprint recognition result back to the video background server.

[0182] Step S307: the video backend server generates an access secondary account instruction corresponding to the business intention.

[0183] Step S308, the video backend server determines whether the corresponding secondary account already exists according to the voiceprint recognition result. If so, it returns the business data corresponding to the secondary account to the client according to the secondary account access instruction; if not, it creates a new secondary account, and the secondary account is a sub-account of the primary account.

[0184] The present application can determine the target secondary identity corresponding to the user identity by identifying the user identity and business intention that generates the current biometric information. Therefore, the business instructions executed based on the target secondary identity not only meet the user's current business intention, but also match the user identity. Furthermore, the present application only needs to collect the user's target biometric information, and can simultaneously determine the user's business intention and user identity. The user does not have to perform two operations of determining the business intention and determining the user identity, thereby reducing the user's operating costs and improving the efficiency of the terminal in executing business instructions that match both the user's business intention and user identity.

[0185] See also Figure 6 , is a flow chart of determining a target user identity and a target secondary identity provided by an embodiment of the present application, wherein determining the target user identity and the target secondary identity includes the following steps S401 to S404, and steps S401 to S404 are the above steps. Figure 3 Specific embodiments of step S102-step S103 in the corresponding embodiment:

[0186] Step S401, converting the target voice data into text data, semantically recognizing the text data, and obtaining the business intent.

[0187] Specifically, when the target biometric information is target voice data, the server divides the target voice data into multiple audio frames according to a preset frame length and a preset frame shift, and there is partial overlap between the audio frames, and the overlapping length is equal to the preset frame shift.

[0188] For example, the target voice data with a time dimension of 0-30ms can be divided into audio frame 1: voice data between 0-20ms and audio frame 2: voice data between 10-30ms according to the frame length of 20ms and the frame shift of 10ms.

[0189] The spectrum parameters of each audio frame are extracted, wherein the spectrum parameters are short-time spectrum feature parameters of the audio frame, and the short-time spectrum feature parameters are parameters extracted based on the physiological structure of the vocal organs such as the glottis, vocal tract or nasal cavity.

[0190] The short-time spectrum characteristic parameters may include at least one of the following parameters: the fundamental frequency spectrum and its contour, the energy of the fundamental frequency frame, the spectrum envelope, the occurrence frequency of the fundamental frequency formant and its trajectory.

[0191] Extract the linear prediction parameters of each audio frame, where the linear prediction parameters are the spectrum fitting feature parameters of the audio frame. From an auditory perspective, the spectrum fitting feature parameters are proposed by simulating the characteristics of the human ear's perception of sound frequency. From a mathematical perspective, the spectrum fitting feature parameters are several "past" audio frames to approximate the current audio frame, and the corresponding approximation parameters are used to estimate the speech features.

[0192] The spectrum fitting characteristic parameters may include at least one of: linear prediction cepstrum (LPCC), line spectral pair (LSP), autocorrelation and logarithmic area ratio, Mel frequency cepstrum (MFCC), perceptual linear prediction (PLP) and the like.

[0193] According to the above method, the spectrum parameters and linear prediction parameters extracted from each audio frame are combined into a vector, so each audio frame can be expressed as a multidimensional vector (also called a feature vector). The acoustic model is used to determine the state of the feature vector corresponding to each audio frame. Generally speaking, the states of adjacent audio frames should be the same, because the frame length of each audio frame is relatively short, at the millisecond level.

[0194] The states corresponding to several audio frames (usually 3 audio frames) are combined into a phoneme. A phoneme is the smallest unit of speech. A phoneme is a unit separated from the perspective of timbre. A phoneme that exists alone or several phonemes combined together are called a syllable.

[0195] Then combine several phonemes into words (or words). Due to the time-varying nature of speech signals, noise and other unstable factors, each word is closely related to the context. In order to further improve the accuracy of speech-to-text conversion, adaptive adjustments must be made based on the context of all words. Therefore, the server can use a language model to combine the recognized words into logical and unambiguous sentences, and obtain text data corresponding to the target speech data.

[0196] The server can obtain an entity-predicate knowledge graph, which includes multiple entity strings and predicate strings, and each entity string (or predicate string) identifies whether the string is an entity attribute or a predicate attribute. The server can use a multi-mode string matching algorithm (the multi-mode string matching algorithm may include: AC automaton, hash function matching, etc.) to perform multi-mode string matching on the text data and the entity-predicate knowledge graph to determine the matching string in the text data and whether the string is an entity attribute or a predicate attribute. The server can use the string belonging to the entity attribute in the text data as an entity and the string belonging to the predicate attribute as a predicate. The entities and predicates identified from the text data are combined into business intents.

[0197] Step S402, calling the identity recognition model corresponding to the first-level identity identifier to determine the matching result between the target voice data and at least one template user identity; the identity recognition model is a classification model generated based on the at least one template user identity and the template voice data corresponding to the at least one template user identity.

[0198] Specifically, the server obtains an identity recognition model corresponding to the first-level identity identifier. The identity recognition model is a classification model trained based on at least one template user identity and template voice data corresponding to each template user identity. The template user identity can be understood as a user identity that the server has created. Each template user identity has a corresponding second-level identity identifier, and the second-level identity identifier is a sub-identifier of the first-level identity identifier.

[0199] The identity recognition model includes a feature generator and a pattern matcher:

[0200] The feature generator is used to divide the target speech data into multiple audio frames, extract the spectrum parameters and linear prediction parameters of each audio frame (the process of extracting the spectrum parameters and linear prediction parameters of each audio frame can be seen in the above step S401), combine the spectrum parameters of all audio frames into the spectrum parameters of the target speech data, and combine the linear prediction parameters of all audio frames into the linear prediction parameters of the target speech data. The spectrum parameters of the target speech data and the linear prediction parameters of the target speech data are combined into the voiceprint features of the target speech data (referred to as the target voiceprint features) in a predetermined order.

[0201] The pattern matcher is used to identify the similarity (or matching probability) between the target voiceprint feature and at least one template voiceprint feature, and take the at least one matching probability obtained as the matching result. The template voiceprint feature is the voiceprint feature of the template speech data (the extraction process of the template voiceprint feature is the same as the extraction process of the target biometric feature).

[0202] Since the template voice data is the voice data corresponding to the template user identity, the similarity (or matching result) between the target biometric feature and at least one template voiceprint feature is equivalent to the matching degree between the target voice data and at least one template user identity.

[0203] The pattern matcher is a model with predictive classification functions, such as a BP (Back Propagation) neural network model, a convolutional neural network model, or various regression models (for example, a linear regression model, a logistic regression model).

[0204] Step S403: If there is a matching result satisfying the matching condition in at least one matching result, the template user identity corresponding to the matching result satisfying the matching condition is used as the target user identity.

[0205] Specifically, the server obtains a preset probability threshold. If the matching result is greater than the preset probability threshold, it means that the matching result is a matching result that meets the matching condition.

[0206] When there is a matching result satisfying the matching condition among the at least one matching result obtained, the template user identity corresponding to the matching result satisfying the matching condition is used as the target user identity.

[0207] Step S404: extracting a target identity corresponding to the target user identity from the secondary identity set corresponding to the at least one template user identity; the secondary identity in the secondary identity set is a sub-identifier of the primary identity.

[0208] Specifically, each template user identity has a corresponding secondary identity identifier, and the secondary identity identifier of each template user identity is a sub-identifier of the primary identity identifier. The secondary identity identifiers of all template user identities can be combined into a secondary identity identifier set.

[0209] The server may extract the secondary identity identifier of the target user identity (ie, the template user identity corresponding to the matching result that satisfies the matching condition) from the above secondary identity identifier set as the target secondary identity identifier.

[0210] For example, the matching probability between the target voiceprint feature of the target voice data and the template voiceprint feature 1 (corresponding to the template user identity 1) is 0.1, the matching probability between the target voiceprint feature and the template voiceprint feature 2 (corresponding to the template user identity 2) is 0.8, and the matching probability between the target voiceprint feature and the template voiceprint feature 3 (corresponding to the template user identity 3) is 0.1. If the preset probability threshold is 0.5, it means that the matching result between the target biometric feature and the template voiceprint feature 2 meets the matching condition, and the server can use the template user identity 2 as the target user identity, and the secondary identity of the template user identity as the target secondary identity.

[0211] Optionally, when there is a matching result satisfying the matching condition among at least one of the obtained matching results, the server may send a play animation instruction to the client, so that the client plays a target animation according to the play animation instruction, and the target animation may be a lightweight animation.

[0212] Subsequently, when the business instruction corresponding to the business intent is executed, the server may send a stop animation playing instruction to the client, so that the client stops the target animation according to the stop animation playing instruction.

[0213] The above steps S403 and S404 describe the situation when there is a matching result satisfying the matching condition in at least one matching result obtained. The following describes the situation when there is no matching result satisfying the matching condition in at least one matching result obtained:

[0214] If the matching result is less than or equal to the preset probability threshold, it is considered that the matching result is a matching result that does not meet the matching condition.

[0215] When there is no matching result that satisfies the matching condition in at least one of the matching results obtained (or when at least one of the matching results obtained does not satisfy the matching condition), the server may create a user identity for the current user (referred to as the target user identity), create a secondary identity identifier for the target user identity (referred to as the target secondary identity identifier), and set the target secondary identity identifier as a sub-identifier of the primary identity identifier.

[0216] The server can also identify the age information corresponding to the target voice data, and search for an image matching the age information from the image material library as the identity avatar.

[0217] The server can store the target user identity, the target secondary identity identifier, and the identity avatar in association.

[0218] Subsequently, the server may use the target user identity as a new template user identity, and add the target secondary user identity identifier to the target secondary identity identifier set.

[0219] For example, the matching probability between the target voiceprint feature of the target voice data and the template voiceprint feature 1 (corresponding to the template user identity 1) is 0.1, the matching probability between the target voiceprint feature and the template voiceprint feature 2 (corresponding to the template user identity 2) is 0.2, and the matching probability between the target voiceprint feature and the template voiceprint feature 3 (corresponding to the template user identity 3) is 0.2. If the preset probability threshold is 0.5, it means that there is no matching result that meets the matching condition among the above three matching results, then the server can re-create the target user identity (for example, user identity 4) and re-create the target secondary identity for the target user identity, and set the re-created target secondary identity as the sub-identity of the primary identity.

[0220] Optionally, the above describes the process of using the identity recognition model. The following describes the process of training the identity recognition model. The training process is explained by taking a template user identity and corresponding template voice data for a model training as an example:

[0221] The server obtains the template voice data of the template user identity and generates a label vector (called an identity label vector) of the template voice data. The identity label vector is used to identify the template user identity to which the template voice data belongs.

[0222] An initial classification model is obtained, and a matching degree between the template voice data and at least one template user identity is predicted based on the initial classification model, and the obtained matching degrees are combined into an identity prediction vector.

[0223] The difference between the identity label vector and the identity prediction vector is determined as a classification error, and the classification error is back-propagated to the initial classification model to adjust model parameters in the initial classification model.

[0224] For example, there are three template user identities (template user identity 1, template user identity 2, and template user identity 3), and the one currently being trained is template user identity 2. The identity label vector of the template speech data of template user identity 2 is: [0, 1, 0]. If the initial classification model predicts that the matching degree between the template speech data of template user identity 2 and template user identity 1 is 0.4, the matching degree between template user identity 2 and template user identity 3 is 0.3, then the identity prediction vector is: [0.4, 0.3, 0.3]. The classification error can be: (0-0.4) 2 +(1-0.3) 2 +(0-0.3) 2 =0.41. The calculated classification error is back-propagated to the initial classification model to adjust the model parameters in the initial classification model.

[0225] The server can continuously train the initial classification model in the above manner. When the number of training times reaches a threshold, or when the change in the model parameters between two consecutive adjustments is small, the trained initial classification model can be used as an identity recognition model.

[0226] From the above, it can be seen that the server may newly create a target user identity and a target secondary identity. The newly created target user identity will be used as a new template user identity. In this case, the server needs to retrain the identity recognition model, and the new identity recognition model will add a new category output based on the original identity recognition model. The newly added category output is used to output the probability that the voice data belongs to the newly added target user identity.

[0227] The present application can determine the target secondary identity corresponding to the user identity by identifying the user identity and business intention that generates the current biometric information. Therefore, the business instructions executed based on the target secondary identity not only meet the user's current business intention, but also match the user identity. Furthermore, the present application only needs to collect the user's target biometric information, and can simultaneously determine the user's business intention and user identity. The user does not have to perform two operations, namely, determining the business intention and determining the user identity, which can reduce the user's operating costs and improve the efficiency of the terminal in executing business instructions that match both the user's business intention and user identity.

[0228] See also Figure 7 , is a flow chart of another data processing method provided in an embodiment of the present application, and the data processing may include the following steps:

[0229] Step S501, the process starts.

[0230] Step S502: The server obtains voice data.

[0231] Specifically, the current first-level account (which can correspond to the first-level identity identifier in this application) has logged in to the client, that is, the first-level account is in a valid state, and the server receives voice data sent by the client, where the voice data is the data collected by the client when the user voice inputs "enter the sub-account" in the client.

[0232] Step S503: The server recognizes that the business intention of the voice data is to access the secondary account, and extracts the target voiceprint features of the voice data based on the identity recognition model. The specific process of extracting the target voiceprint features can be found in the above Figure 6 This corresponds to step S402 in the embodiment.

[0233] Step S504: the server performs pattern matching between the extracted target voiceprint features and the existing template voiceprint features.

[0234] Step S505, the server determines whether there is a template voiceprint feature matching the target biometric feature in the existing template voiceprint features based on the pattern matching result. If so, execute steps S507-S508; if not, execute steps S506 and S508.

[0235] Step S506, the server creates a new secondary account (which may correspond to the target secondary identity in this application) according to the secondary account access intention, associates the secondary account with the extracted target voiceprint feature, and logs the newly created secondary account into the client.

[0236] Step S507, the server searches for a secondary account (which may correspond to the target secondary identity in this application) corresponding to the matching template voiceprint feature. The secondary account is an existing secondary account, and the server returns the service data under the secondary account to the client.

[0237] Step S508, end the process.

[0238] When the client in the above steps is a video client and the video client is installed in a smart TV, each family member sharing the smart TV uniquely corresponds to a user identity and a secondary account (which can correspond to the secondary identity identifier in this application) through the voiceprint feature. The server can determine the viewing history, the movies they follow, and the voiceprint features of each family member based on the unique secondary account, thereby achieving personalized recommendations.

[0239] The following scenario is explained by taking the case where user A has created a secondary account of the client, but user B has not yet created a secondary account as an example: the client collects the voice data input by the user to "enter the sub-account".

[0240] Please see further Figure 8 , is a timing diagram of another data processing method provided in an embodiment of the present application, the data processing method comprises the following steps:

[0241] Step S601: The client collects voice data of user A.

[0242] Specifically, the current primary account logs in to the client in the terminal device, that is, the primary account is in a valid state, and the user inputs "login to the sub-account" into the client by voice, and the client collects the voice data of the user's voice input "enter the sub-account".

[0243] Step S602: The client sends the voice data to the server.

[0244] In step S603, the server determines the business intent through semantic recognition, and matches the corresponding secondary account through voiceprint recognition, logs in the secondary account to the client, and searches for the behavior data under the secondary account (for example, user A's historical viewing records, videos followed, commented videos, and search records, etc.), and generates recommendation data based on the behavior data.

[0245] Step S604: The server returns the recommendation data to the client.

[0246] Step S605, the client collects the voice data of user B: "Log in to the sub-account".

[0247] Step S606: The client uploads the voice data of user B to the server.

[0248] Step S607: the server determines the business intent through semantic recognition, and if no corresponding secondary account is matched through voiceprint recognition, a new secondary account is created, and the new secondary account is used as a sub-account of the primary account, and the new secondary account is logged in to the client.

[0249] Step S608: User B generates viewing behavior data (watched videos, followed videos, commented videos, search history, etc.) based on the newly created secondary account in the client.

[0250] Step S609: the client uploads the movie-watching behavior data of user B to the server.

[0251] In step S610, the server associates and stores the movie-watching behavior data with the newly created secondary account for subsequent generation of personalized recommendation data for user B.

[0252] For further information, see Fig. 9 , is a schematic diagram of the structure of a data processing device provided in an embodiment of the present application. Fig. 9 As shown, the data processing device 1 can be applied to the above Figure 3-Figure 8 Corresponding to the server in the embodiment, the data processing device 1 may include: a first acquisition module 11 , an identification module 12 , a second acquisition module 13 and a determination module 14 .

[0253] The first acquisition module 11 is used to acquire target biological information when the primary identity identifier is in a valid state;

[0254] An identification module 12, used to identify the business intention and target user identity corresponding to the target biometric information;

[0255] The second acquisition module 13 is used to acquire a target secondary identity identifier corresponding to the target user identity; the target secondary identity identifier is a sub-identity identifier of the primary identity identifier;

[0256] The determination module 14 is used to execute a business instruction corresponding to the business intention based on the target secondary identity.

[0257] The specific functional implementation of the first acquisition module 11, the identification module 12, the second acquisition module 13 and the determination module 14 can be referred to above. Figure 3 Steps S101 to S104 in the corresponding embodiment will not be described in detail here.

[0258] See also Fig. 9 , the target biological information includes target voice data;

[0259] The identification module 12 may include: a conversion unit 121 , a calling unit 122 and a first determination unit 123 .

[0260] A conversion unit 121, configured to convert the target voice data into text data, semantically identify the text data, and obtain the business intent;

[0261] A calling unit 122 is used to call the identity recognition model corresponding to the primary identity identifier to determine the matching result between the target voice data and at least one template user identity; the identity recognition model is a classification model generated according to the at least one template user identity and the template voice data corresponding to the at least one template user identity;

[0262] A first determining unit 123 is configured to use, if there is a matching result satisfying the matching condition in at least one matching result, the template user identity corresponding to the matching result satisfying the matching condition as the target user identity;

[0263] Then the second acquisition module 13 may include: a first extraction unit 131 .

[0264] The first extraction unit 131 is used to extract a target identity corresponding to the target user identity from a set of secondary identity corresponding to the at least one template user identity; the secondary identity in the set of secondary identity is a sub-identifier of the primary identity.

[0265] The identification module 12 may further include: a second determination unit 124 .

[0266] The second determining unit 124 is configured to create the target user identity, identify the age information corresponding to the target voice data, and search for an identity avatar matching the age information in an image material library if there is no matching result satisfying the matching condition in the at least one matching result;

[0267] The second acquisition module 13 may include: a second extraction unit 132 .

[0268] The second extraction unit 132 is used to create the target secondary user identifier for the target user identity, set the target secondary user identifier as a sub-identifier of the primary identity identifier, and associate and store the target user identity, the target secondary identity identifier, and the identity avatar.

[0269] The specific processes of the conversion unit 121, the calling unit 122, the first determination unit 123, the second determination unit 124, the first extraction unit 131 and the second extraction unit 132 can be referred to above. Figure 6 Steps S401 to S404 in the corresponding embodiment are not described in detail here.

[0270] When the first determination unit 123 and the first extraction unit 131 determine the target user identity and the target secondary identity, the second determination unit 124 and the second extraction unit 132 no longer execute the corresponding steps; when the second determination unit 124 and the second extraction unit 132 determine the target user identity and the target secondary identity, the first determination unit 123 and the first extraction unit 131 no longer execute the corresponding steps.

[0271] See also Fig. 9 , the identity recognition model includes a feature generator and a pattern matcher;

[0272] The calling unit 122 may include: an extracting subunit 1221 and a matching subunit 1222 .

[0273] An extraction subunit 1221, configured to extract a target voiceprint feature of the target speech data based on the feature generator;

[0274] The matching subunit 1222 is used to determine the matching probability between the target voiceprint feature and at least one template voiceprint feature based on the pattern matcher, and use the obtained matching probabilities as matching results; the at least one template voiceprint feature is the voiceprint feature corresponding to the at least one template voice data respectively;

[0275] The extraction subunit 1221 is specifically used to extract the spectrum parameters and linear prediction parameters of the target speech data based on the feature generator, and obtain the target voiceprint features according to the spectrum parameters and the linear prediction parameters; the spectrum parameters are short-time spectrum feature parameters of the target speech data; the linear prediction parameters are spectrum fitting feature parameters of the target speech data.

[0276] The specific process of the extraction subunit 1221 and the matching subunit 1222 can be found in the above Figure 6 The step S402 in the corresponding embodiment will not be described in detail here.

[0277] See also Fig. 9 , the business intention includes the client secondary login object switching intention;

[0278] The determination module 14 includes: a first generation unit 141 .

[0279] The first generating unit 141 is used to generate a switching instruction corresponding to the secondary login object switching intention of the client, and according to the switching instruction, the target secondary identity is used as the secondary login object of the client; the switching instruction belongs to the business instruction.

[0280] The specific process of the first generation unit 141 can be found in the above Figure 3 The step S104 in the corresponding embodiment will not be described in detail here.

[0281] See also Fig. 9 , the business intention includes business data query intention;

[0282] The determining module 14 may include: a second generating unit 142 .

[0283] The second generating unit 142 is used to generate a query instruction corresponding to the business data query intention, query the target business data corresponding to the target secondary identity, and return the target business data to the client; the query instruction belongs to the business instruction.

[0284] The specific process of the second generation unit 142 can be found in the above Figure 3 The step S104 in the corresponding embodiment will not be described in detail here.

[0285] See also Fig. 9 The data processing device 1 may include: a first acquisition module 11, an identification module 12, a second acquisition module 13 and a determination module 14, and may also include: a storage module 15, a training module 16 and a playback module 17.

[0286] The storage module 15 is used to obtain the behavior data of the user corresponding to the target secondary identity in the client, and store the behavior data and the target secondary identity in association; the behavior data is used to generate recommended business data for the user.

[0287] The training module 16 is used to obtain template voice data corresponding to the template user identity, generate an identity label vector corresponding to the template voice data, obtain an initial classification model, predict the matching degree between the template voice data and the at least one template user identity based on the initial classification model, obtain an identity prediction vector according to the obtained matching degree, determine the classification error according to the identity label vector and the identity prediction vector, train the initial classification model according to the classification error, and obtain the identity recognition model.

[0288] A playing module 17, configured to send an animation playing instruction to a client when there is a matching result satisfying the matching condition in the at least one matching result, instructing the client to play a target animation;

[0289] The playing module 17 is further configured to send a stop animation playing instruction to the client when the business instruction is executed, instructing the client to close the target animation.

[0290] The specific processes of the storage module 15, the training module 16 and the playback module 17 can be found in the above Figure 6 The step S404 in the corresponding embodiment will not be described in detail here.

[0291] For further information, see Fig.10 , is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 3-Figure 8 The server in the corresponding embodiment may be an electronic device 1000, such as Fig.10As shown, the electronic device 1000 may include: a user interface 1002, a processor 1004, an encoder 1006 and a memory 1008. The signal receiver 1016 is used to receive or send data via a cellular interface 1010, a WIFI interface 1012, ..., or an NFC interface 1014. The encoder 1006 encodes the received data into a data format for computer processing. A computer program is stored in the memory 1008, and the processor 1004 is configured to perform the steps in any of the above method embodiments through the computer program. The memory 1008 may include a volatile memory (e.g., a dynamic random access memory DRAM), and may also include a non-volatile memory (e.g., a one-time programmable read-only memory OTPROM). In some instances, the memory 1008 may further include a memory remotely arranged relative to the processor 1004, and these remote memories may be connected to the electronic device 1000 via a network. The user interface 1002 may include: a keyboard 1018 and a display 1020.

[0292] exist Fig.10 In the electronic device 1000 shown, the processor 1004 may be used to call a computer program stored in the memory 1008 to implement:

[0293] When the primary identity is in a valid state, the target biological information is obtained;

[0294] Identifying the business intent and target user identity corresponding to the target biometric information;

[0295] Acquire a target secondary identity identifier corresponding to the target user identity; the target secondary identity identifier is a sub-identifier of the primary identity identifier;

[0296] Execute a business instruction corresponding to the business intention based on the target secondary identity.

[0297] It should be understood that the electronic device 1000 described in the embodiment of the present invention can execute the above Figures 3 to 8 The description of the data processing method in the corresponding embodiment can also be performed as described above. Fig. 9 The description of the data processing device 1 in the corresponding embodiment will not be repeated here. In addition, the description of the beneficial effects of the same method will not be repeated here either.

[0298] In addition, it should be pointed out that: an embodiment of the present invention further provides a computer storage medium, and the computer storage medium stores a computer program executed by the data processing device 1 mentioned above, and the computer program includes program instructions. When the processor executes the program instructions, it can execute the above-mentioned Figures 3 to 8The description of the data processing method in the corresponding embodiment will not be repeated here. In addition, the description of the beneficial effects of the same method will not be repeated. For technical details not disclosed in the computer storage medium embodiment involved in the present invention, please refer to the description of the method embodiment of the present invention.

[0299] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0300] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A data processing method, characterized in that: include: When the primary identity is in a valid state, obtaining target biometric information, wherein the target biometric information includes target voice data; Convert the target voice data into text data, semantically identify the text data, and obtain the business intent; Invoking an identity recognition model corresponding to the primary identity identifier to determine a matching result between the target voice data and at least one template user identity; The identity recognition model is a classification model generated according to the at least one template user identity and the template voice data respectively corresponding to the at least one template user identity; If there is a matching result satisfying the matching condition in at least one matching result, taking the template user identity corresponding to the matching result satisfying the matching condition as the target user identity, sending a play animation instruction to the client, instructing the client to play the target animation; Obtaining a target secondary identity identifier corresponding to the target user identity; The target secondary identity identifier is a sub-identity identifier of the primary identity identifier; Executing a business instruction corresponding to the business intention based on the target secondary identity; When the execution of the service instruction is completed, a stop animation playing instruction is sent to the client, instructing the client to close the target animation.

2. The method according to claim 1, characterized in that The step of obtaining a target secondary identity corresponding to the target user identity includes: A target identity corresponding to the target user identity is extracted from a set of secondary identity identifiers corresponding to the at least one template user identity; the secondary identity identifiers in the set of secondary identity identifiers are sub-identifiers of the primary identity identifier.

3. The method according to claim 2, characterized in that Also includes: If there is no matching result satisfying the matching condition in the at least one matching result, creating the target user identity; Identify age information corresponding to the target voice data, and search for an identity head image matching the age information in an image material library; The step of obtaining a target secondary identity corresponding to the target user identity includes: Creating the target secondary user identifier for the target user identity; Setting the target secondary user identifier as a sub-identifier of the primary identity identifier; The target user identity, the target secondary identity identifier and the identity avatar are stored in association.

4. The method according to claim 2, characterized in that: The identity recognition model includes a feature generator and a pattern matcher; The calling of the identity recognition model corresponding to the primary identity identifier to determine a matching result between the target voice data and at least one template user identity includes: Based on the feature generator, extracting target voiceprint features of the target voice data; Based on the pattern matcher, the matching probability between the target voiceprint feature and at least one template voiceprint feature is determined, and the obtained matching probabilities are all used as matching results; the at least one template voiceprint feature is the voiceprint feature corresponding to the at least one template voice data respectively.

5. The method according to claim 4, characterized in that The step of extracting the target voiceprint feature of the target voice data based on the feature generator includes: Based on the feature generator, extract the spectrum parameters and linear prediction parameters of the target speech data; the spectrum parameters are the short-time spectrum feature parameters of the target speech data; the linear prediction parameters are the spectrum fitting feature parameters of the target speech data; The target voiceprint feature is obtained according to the spectrum parameter and the linear prediction parameter.

6. The method according to claim 2, characterized in that Also includes: Obtain template voice data corresponding to the template user identity; Generating an identity tag vector corresponding to the template speech data; Acquire an initial classification model, predict a matching degree between the template voice data and the at least one template user identity based on the initial classification model, and obtain an identity prediction vector according to the acquired matching degree; A classification error is determined according to the identity label vector and the identity prediction vector, and the initial classification model is trained according to the classification error to obtain the identity recognition model.

7. The method according to claim 1, characterized in that The business intention includes the client secondary login object switching intention; The executing the business instruction corresponding to the business intention based on the target secondary identity includes: Generate a switching instruction corresponding to the secondary login object switching intention of the client; the switching instruction belongs to the business instruction; According to the switching instruction, the target secondary identity is used as a secondary login object of the client.

8. The method according to claim 7, characterized in that Also includes: Obtaining behavior data of a user corresponding to the target secondary identity in the client; The behavior data is used to generate recommended business data for the user; The behavior data and the target secondary identity are associated and stored.

9. The method according to claim 1, characterized in that: The business intention includes business data query intention; The executing the business instruction corresponding to the business intention based on the target secondary identity includes: Generate a query instruction corresponding to the business data query intention; the query instruction belongs to the business instruction; Query the target business data corresponding to the target secondary identity identifier, and return the target business data to the client.

10. The method according to claim 1, characterized in that The user rights of the target secondary identity are the same as the user rights of the primary identity.

11. A data processing device, characterized in that: include: A first acquisition module, used for acquiring target biometric information when the primary identity identifier is in a valid state, wherein the target biometric information includes target voice data; A recognition module, used to convert the target voice data into text data, semantically recognize the text data, and obtain the business intent; Invoking an identity recognition model corresponding to the primary identity identifier to determine a matching result between the target voice data and at least one template user identity; The identity recognition model is a classification model generated according to the at least one template user identity and the template voice data respectively corresponding to the at least one template user identity; If there is a matching result satisfying the matching condition in at least one matching result, taking the template user identity corresponding to the matching result satisfying the matching condition as the target user identity, sending a play animation instruction to the client, instructing the client to play the target animation; A second acquisition module is used to acquire a target secondary identity corresponding to the target user identity; The target secondary identity identifier is a sub-identity identifier of the primary identity identifier; The determination module is used to execute a business instruction corresponding to the business intention based on the target secondary identity identifier; when the business instruction is executed, a stop animation playback instruction is sent to the client to instruct the client to close the target animation.

12. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 10.

13. A computer storage medium, characterized in that: The computer storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the method according to any one of claims 1 to 10 is executed.

Images