CN113889120A

CN113889120A - Voiceprint feature extraction method and device, electronic equipment and storage medium

Info

Publication number: CN113889120A
Application number: CN202111143893.1A
Authority: CN
Inventors: 赵情恩
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2021-09-28
Filing date: 2021-09-28
Publication date: 2022-01-04

Abstract

The disclosure provides a voiceprint feature extraction method and device, electronic equipment and a storage medium, relates to the technical field of artificial intelligence, and particularly relates to a voice recognition technology. The implementation scheme is as follows: a voiceprint feature extraction method comprises the following steps: acquiring initial voiceprint characteristic data about a speaker; generating an initial feature vector of the speaker based on the initial voiceprint feature data; generating a covariance matrix corresponding to the initial eigenvector; generating an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to Gaussian distribution; and extracting voiceprint features of the speaker based on the updated feature vector.

Description

Voiceprint feature extraction method and device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of artificial intelligence technologies, and in particular, to a voiceprint feature extraction method and apparatus, an electronic device, and a storage medium.

Background

Artificial intelligence is the subject of research that makes computers simulate some human thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.), both at the hardware level and at the software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like; the artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, machine learning/deep learning, a big data processing technology, a knowledge map technology and the like.

In speech recognition technology, voiceprint feature extraction is crucial. Voiceprint feature extraction may generally include various modules for extracting front-end features (also referred to as low-level features), for extracting features of a speaker (also referred to as high-level features), and for back-end classification. The voice data can obtain corresponding recognition results after being processed by the modules.

However, in the process of extracting the voiceprint features, the voiceprint features are often interfered by various factors such as the speaker and the environment, and the accuracy of extracting the voiceprint features is influenced to a certain extent. How to eliminate these influences and extract the voiceprint features of the speaker more accurately has become a popular research field in the voiceprint feature extraction technology in recent years.

The approaches described in this section are not necessarily approaches that have been previously conceived or pursued. Unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section. Similarly, unless otherwise indicated, the problems mentioned in this section should not be considered as having been acknowledged in any prior art.

Disclosure of Invention

The disclosure provides a voiceprint feature extraction method and device, electronic equipment and a storage medium.

According to an aspect of the present disclosure, there is provided a voiceprint feature extraction method including: acquiring initial voiceprint characteristic data about a speaker; generating an initial feature vector of the speaker based on the initial voiceprint feature data; generating a covariance matrix corresponding to the initial eigenvector; generating an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to Gaussian distribution; and extracting voiceprint features of the speaker based on the updated feature vector.

According to another aspect of the present disclosure, there is provided a method for training a voiceprint feature extraction model, comprising: providing sample initial voiceprint characteristic data about a predetermined speaker; generating a sample initial feature vector of a predetermined speaker based on the sample initial voiceprint feature data; generating a sample covariance matrix corresponding to the sample initial feature vector; generating an updated sample eigenvector of the predetermined speaker based on the sample initial eigenvector and the sample covariance matrix, wherein the updated sample eigenvector is a posterior mean vector of the sample initial eigenvector according to Gaussian distribution; extracting voiceprint features of the predetermined speaker based on the updated sample feature vector; and acquiring network parameters for updating the voiceprint feature extraction model based on the voiceprint features to train the voiceprint feature extraction model.

According to another aspect of the present disclosure, there is provided a voiceprint feature extraction apparatus including: an acquisition unit configured to acquire initial voiceprint feature data about a speaker; a first generating unit configured to generate an initial feature vector of the speaker based on the initial voiceprint feature data; a second generation unit configured to generate a covariance matrix corresponding to the initial eigenvector; a third generating unit configured to generate an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to a gaussian distribution; and an extraction unit configured to extract voiceprint features of the speaker based on the updated feature vector.

According to another aspect of the present disclosure, there is provided an apparatus for training a voiceprint feature extraction model, comprising: a providing unit configured to provide sample initial voiceprint feature data about a predetermined speaker; a first sample generation unit configured to generate a sample initial feature vector of a predetermined speaker based on the sample initial voiceprint feature data; a second sample generation unit configured to generate a sample covariance matrix corresponding to the sample initial feature vector; a third sample generation unit configured to generate an updated sample feature vector of the predetermined speaker based on the sample initial feature vector and the sample covariance matrix, wherein the updated sample feature vector is a posterior mean vector of the sample initial feature vector according to a gaussian distribution; a sample extraction unit configured to extract voiceprint features of a predetermined speaker based on the updated sample feature vector; and a network parameter acquisition unit configured to acquire a network parameter for updating the voiceprint feature extraction model based on the voiceprint feature to train the voiceprint feature extraction model.

According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor, which when executed by the at least one processor, cause the at least one processor to perform the method as described above.

According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method as described above.

According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein the computer program realizes the method as described above when executed by a processor.

According to one or more embodiments of the present disclosure, a voiceprint feature of a speaker can be accurately extracted.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the embodiments and, together with the description, serve to explain the exemplary implementations of the embodiments. The illustrated embodiments are for purposes of illustration only and do not limit the scope of the claims. Throughout the drawings, identical reference numbers designate similar, but not necessarily identical, elements.

Fig. 1 illustrates a schematic diagram of an exemplary system in which various methods and apparatus described herein may be implemented in accordance with embodiments of the present disclosure.

Fig. 2 shows a flow chart of a voiceprint feature extraction method according to an embodiment of the present disclosure.

FIG. 3 shows a flow diagram of a method for training a voiceprint feature extraction model in accordance with an embodiment of the present disclosure.

FIG. 4 shows a schematic diagram of a voiceprint extraction model according to an embodiment of the disclosure.

Fig. 5 shows a block diagram of a voiceprint feature extraction apparatus according to one embodiment of the present disclosure.

Fig. 6 shows a block diagram of a voiceprint feature extraction apparatus according to another embodiment of the present disclosure.

FIG. 7 shows a block diagram of an apparatus for training a voiceprint feature extraction model according to an embodiment of the present disclosure.

Fig. 8 shows a block diagram of an electronic device to which the embodiments of the present disclosure can be applied.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, the timing relationship, or the importance relationship of the elements, and such terms are used only to distinguish one element from another. In some examples, a first element and a second element may refer to the same instance of the element, and in some cases, based on the context, they may also refer to different instances.

The terminology used in the description of the various described examples in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the elements may be one or more. Furthermore, the term "and/or" as used in this disclosure is intended to encompass any and all possible combinations of the listed items.

In the related art, in order to extract the features of the speaker, the method used may generally include unsupervised Gaussian Mixture Model (GMM), and supervised Deep Neural Network (DNN), Convolutional Neural Network (CNN), etc. The voiceprint feature extraction effect brought by the method is relatively limited, and a bottleneck exists for improving the accuracy of voiceprint feature extraction.

In view of the above problems, according to an aspect of the present disclosure, a voiceprint feature extraction method is provided. Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

Fig. 1 illustrates a schematic diagram of an exemplary system 100 in which various methods and apparatus described herein may be implemented in accordance with embodiments of the present disclosure. Referring to fig. 1, the system 100 includes one or

more client devices

101, 102, 103, 104, 105, and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120.

Client devices

101, 102, 103, 104, 105, and 106 may be configured to execute one or more applications.

In embodiments of the present disclosure, the server 120 may run one or more services or software applications that enable the voiceprint feature extraction methods of the present disclosure to be performed.

In some embodiments, the server 120 may also provide other services or software applications that may include non-virtual environments and virtual environments. In certain embodiments, these services may be provided as web-based services or cloud services, for example, provided to users of

client devices

101, 102, 103, 104, 105, and/or 106 under a software as a service (SaaS) model.

In the configuration shown in fig. 1, server 120 may include one or more components that implement the functions performed by server 120. These components may include software components, hardware components, or a combination thereof, which may be executed by one or more processors. A user operating a

client device

101, 102, 103, 104, 105, and/or 106 may, in turn, utilize one or more client applications to interact with the server 120 to take advantage of the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from system 100. Accordingly, fig. 1 is one example of a system for implementing the various methods described herein and is not intended to be limiting.

The user may use

client devices

101, 102, 103, 104, 105, and/or 106 to generate voice data for the voiceprint feature extraction method of the present disclosure. The client device may provide an interface that enables a user of the client device to interact with the client device. The client device may also output information to the user via the interface. Although fig. 1 depicts only six client devices, those skilled in the art will appreciate that any number of client devices may be supported by the present disclosure.

Client devices

101, 102, 103, 104, 105, and/or 106 may include various types of computer devices, such as portable handheld devices, general purpose computers (such as personal computers and laptop computers), workstation computers, wearable devices, gaming systems, thin clients, various messaging devices, sensors or other sensing devices, and so forth. These computer devices may run various types and versions of software applications and operating systems, such as MICROSOFT Windows, APPLE iOS, UNIX-like operating systems, Linux, or Linux-like operating systems (e.g., GOOGLE Chrome OS); or include various Mobile operating systems such as MICROSOFT Windows Mobile OS, iOS, Windows Phone, Android. Portable handheld devices may include cellular telephones, smart phones, tablets, Personal Digital Assistants (PDAs), and the like. Wearable devices may include head mounted displays and other devices. The gaming system may include a variety of handheld gaming devices, internet-enabled gaming devices, and the like. The client device is capable of executing a variety of different applications, such as various Internet-related applications, communication applications (e.g., email applications), Short Message Service (SMS) applications, and may use a variety of communication protocols.

Network 110 may be any type of network known to those skilled in the art that may support data communications using any of a variety of available protocols, including but not limited to TCP/IP, SNA, IPX, etc. By way of example only, one or more networks 110 may be a Local Area Network (LAN), an ethernet-based network, a token ring, a Wide Area Network (WAN), the internet, a virtual network, a Virtual Private Network (VPN), an intranet, an extranet, a Public Switched Telephone Network (PSTN), an infrared network, a wireless network (e.g., bluetooth, WIFI), and/or any combination of these and/or other networks.

The server 120 may include one or more general purpose computers, special purpose server computers (e.g., PC (personal computer) servers, UNIX servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and/or combination. The server 120 may include one or more virtual machines running a virtual operating system, or other computing architecture involving virtualization (e.g., one or more flexible pools of logical storage that may be virtualized to maintain virtual storage for the server). In various embodiments, the server 120 may run one or more services or software applications that provide the functionality described below.

The computing units in server 120 may run one or more operating systems including any of the operating systems described above, as well as any commercially available server operating systems. The server 120 may also run any of a variety of additional server applications and/or middle tier applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, and the like.

In some implementations, the server 120 may include one or more applications to analyze and consolidate data feeds and/or event updates received from users of the

client devices

101, 102, 103, 104, 105, and 106. Server 120 may also include one or more applications to display data feeds and/or real-time events via one or more display devices of

client devices

101, 102, 103, 104, 105, and 106.

In some embodiments, the server 120 may be a server of a distributed system, or a server incorporating a blockchain. The server 120 may also be a cloud server, or a smart cloud computing server or a smart cloud host with artificial intelligence technology. The cloud Server is a host product in a cloud computing service system, and is used for solving the defects of high management difficulty and weak service expansibility in the traditional physical host and Virtual Private Server (VPS) service.

The system 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store information such as audio files and video files. The database 130 may reside in various locations. For example, the database used by the server 120 may be local to the server 120, or may be remote from the server 120 and may communicate with the server 120 via a network-based or dedicated connection. The database 130 may be of different types. In some embodiments, the database used by the server 120 may be, for example, a relational database. One or more of these databases may store, update, and retrieve data to and from the database in response to the command.

In some embodiments, one or more of the databases 130 may also be used by applications to store application data. The databases used by the application may be different types of databases, such as key-value stores, object stores, or regular stores supported by a file system.

The system 100 of fig. 1 may be configured and operated in various ways to enable application of the various methods and apparatus described in accordance with the present disclosure.

FIG. 2 shows a flow diagram of a voiceprint feature extraction method 200 according to an embodiment of the disclosure. As shown in fig. 2, the method 200 may include the steps of:

s202, acquiring initial voiceprint characteristic data of a speaker;

s204, generating an initial feature vector of the speaker based on the initial voiceprint feature data;

s206, generating a covariance matrix corresponding to the initial feature vector;

s208, generating an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to Gaussian distribution; and

and S210, extracting the voiceprint features of the speaker based on the updated feature vector.

According to the voiceprint feature extraction method disclosed by the disclosure, in order to more accurately extract the voiceprint feature of the speaker, on the basis of generating the initial feature vector of the speaker, a posterior mean vector of the initial feature vector according to gaussian distribution can be obtained by using a covariance matrix capable of representing the correlation of the initial feature vector between different feature dimensions (for example, corresponding feature dimensions related to physiological characteristics of the mouth, vocal cords and the like of the speaker, corresponding feature dimensions related to the environment and the background during speaking and the like). The posterior mean vector can contain richer voiceprint feature information so that the voiceprint features of the speaker can be extracted as much and sufficiently as possible. Therefore, the posterior mean vector is used as an updated feature vector on the basis of the initial feature vector, and can be used for accurately extracting the voiceprint features of the speaker, so that the accuracy of voiceprint feature extraction is improved.

The respective steps of the voiceprint feature extraction method according to the present disclosure will be described in detail below.

In step S202, the speaker may refer to any speaker whose voiceprint features are to be extracted. Initial voiceprint feature data can be obtained for the speaker.

According to some embodiments, the initial voiceprint feature data can be extracted from audio data about the speaker, and the initial voiceprint feature data can include a plurality of sub-feature data having a contextual relationship, wherein the plurality of sub-feature data can correspond to a plurality of frames of audio data.

Specifically, audio data of a speaker may be obtained and framed. For example, a segment of audio data of a speaker may be framed into T frames of audio data (T is a natural number greater than 1), and corresponding voiceprint feature data may be extracted from each frame of audio data in the T frames of audio data to obtain T pieces of voiceprint feature data. Accordingly, the T voiceprint feature data generally correspond to the initial voiceprint feature data for the speaker, and each voiceprint feature data corresponds to one sub-feature data.

Here, in order to be able to obtain information related to the time variation of the voiceprint feature of the speaker, it is considered to frame the audio data into a plurality of audio frames, i.e., T is greater than or equal to 2. Accordingly, the T audio frames have a temporal context relationship therebetween, and therefore the feature data extracted therefrom, i.e., the T sub-feature data, also have a temporal context relationship therebetween. For example, the T sub-feature data can reflect the voiceprint features of the speaker at several consecutive time points (i.e., T frames), and such voiceprint features may be different for different speakers and can be characteristic voiceprint features.

In this way, information related to the time variation of the voiceprint features of the speaker can be included in the voiceprint extraction process, so that more information of the voiceprint features can be extracted conveniently, and the accuracy of voiceprint feature extraction is improved.

After the audio data is obtained, various preprocessing operations may also be performed, including noise removal (e.g., environmental noise, busy tone, polyphonic ringtone, etc.), data enhancement (e.g., aliasing echo, rate of change (e.g., speech rate is faster or slower), time domain and frequency domain random masking, etc.).

Extraction of initial voiceprint feature data from audio data can be performed by various known voiceprint feature extraction techniques, such as MFCC (mel frequency cepstral coefficients), Fbank (filterbank), PLP (perceptual linear prediction), etc. In addition, feature mean regularization (mean reduction) may also be performed.

In the technical scheme of the present disclosure, the processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the related users all conform to the regulations of the related laws and regulations, and do not violate the good custom of the public order.

According to some embodiments, in step S204, the initial voiceprint feature data may be input to a first neural network to obtain an initial feature vector, wherein the initial feature vector includes a plurality of first elements having a temporal context corresponding to a plurality of sub-feature data.

The first neural network may be various types of neural networks known, for example, TDNN (time delay neural network). Because the neural network can be fitted with any distribution, the distribution required to be met does not need to be preset, and a corresponding calculation result is obtained by depending on the learning capability of the neural network. By means of multiple mappings of the neural network on the feature data, the accuracy of calculating the initial feature vector can be improved.

The first neural network may comprise, for example, a multi-layer TDNN (with a ReLU (modified linear unit) activation function) and a fully-connected layer (with a ReLU activation function). In practical applications, for the selection of the first neural network, an appropriate neural network may be selected in consideration of the trade-off between the effect and the efficiency.

Accordingly, the initial feature vector extracted from the initial voiceprint feature data may include T first elements corresponding to the T sub-feature data, and the T first elements may also have a temporal context relationship.

According to some embodiments, at step S206, the initial feature vector may be input to a second neural network to obtain a covariance matrix corresponding to the initial feature vector, wherein the covariance matrix has a plurality of second elements corresponding to a plurality of first elements of the initial feature vector, and each second element characterizes a correlation between the corresponding first elements in different feature dimensions.

The second neural network may also be a neural network of various types known. For example, the second neural network may include two fully connected layers (with a ReLU activation function). The number of layers of the second neural network may be appropriately selected according to the actual situation, for example, 2 to 5 layers.

The second neural network is also referred to herein as the auxiliary fully-connected layer because it takes the initial feature vector as input, while auxiliarily outputting the covariance matrix in preparation for a subsequent application of the normalized exponential function (Softmax).

Accordingly, the covariance matrix may have T second elements corresponding to T first elements of the initial feature vector. Since covariance can be used to represent the correlation between data of different dimensions, each second element in the covariance matrix can characterize the correlation between different feature dimensions of the corresponding first element in the initial feature vector. Here, the different feature dimensions of the initial feature vector may refer to, for example, corresponding feature dimensions related to physiological characteristics of the mouth, vocal cords, and the like of the speaker itself, corresponding feature dimensions related to the environment and the background when speaking, and the like. For example, each second element in the covariance matrix may characterize a correlation between the vocal features on the mouth and vocal features on the vocal cords of the corresponding first element in the initial feature vector.

In this way, by using the covariance matrix of the initial feature vector, the relationship between different feature dimensions of the initial feature vector can be established, and not only the feature information on each single feature dimension is limited, so that more abundant feature information about the speaker can be obtained.

In one example, subsequent calculations related to the covariance matrix may be logarithmically performed after the covariance matrix is obtained, in order to convert the multiplication calculation into the addition calculation, thereby reducing the amount of calculation.

According to some embodiments, at step S208, Softmax (normalized exponential function) may be applied to the covariance matrix to obtain Softmax values for the covariance matrix; and the Softmax value may be multiplied with the initial feature vector to obtain an updated feature vector.

Since Softmax is similar to the normalization operation, after applying Softmax to the covariance matrix, the obtained Softmax value is multiplied by the initial eigenvector, that is, weighted average is performed, so that a posterior average vector of the initial eigenvector according to gaussian distribution can be obtained.

Here, the posterior mean vector can be understood as an accurate value obtained based on actual voiceprint feature data of the speaker. In other words, in the case where the voiceprint feature of the speaker is assumed to conform to the gaussian distribution, the prior probability alone cannot reflect how the specific form of the gaussian distribution is. Under the condition of obtaining the posterior mean vector, the voiceprint features of the speaker can be accurately expressed in Gaussian distribution, so that the voiceprint features of the speaker can be conveniently and accurately extracted.

In one example, to obtain the a posteriori mean vector, a priori mean and covariance may also be initialized (e.g., to 0) to add to the weighted average calculation as above.

In addition, in the case where the initial voiceprint feature data includes a plurality of sub-feature data having a temporal context, the above-described process of obtaining an updated feature vector may represent temporally aggregating a plurality of first elements having a temporal context in the initial feature vector in order to obtain more abundant feature information about the speaker.

According to some embodiments, the updated feature vectors may be mapped by an embedding operation to generate voiceprint features of the speaker at step S210. In this way, the voiceprint features of the speaker can be accurately extracted based on the updated feature vector containing more feature information.

For example, the updated feature vector may be mapped by inputting it into two fully-connected layers (with the ReLU activation function) to generate the speaker's voiceprint features.

As described above, according to the voiceprint feature extraction method of the present disclosure, in order to more accurately extract the voiceprint feature of the speaker, based on the generation of the initial feature vector of the speaker, the posterior mean vector of the initial feature vector according to the gaussian distribution can be obtained by using the covariance matrix capable of representing the correlation between the initial feature vector and different feature dimensions (for example, the corresponding feature dimension related to the physiological characteristics of the oral cavity, vocal cords, etc. of the speaker itself, the corresponding feature dimension related to the environment and the background during speaking, etc.). The posterior mean vector can contain richer voiceprint feature information so that the voiceprint features of the speaker can be extracted as much and sufficiently as possible. Therefore, the posterior mean vector is used as an updated feature vector on the basis of the initial feature vector, and can be used for accurately extracting the voiceprint features of the speaker, so that the accuracy of voiceprint feature extraction is improved.

According to an embodiment of the present disclosure, there is also provided a method for training a voiceprint feature extraction model.

FIG. 3 shows a flow diagram of a method 300 for training a voiceprint feature extraction model in accordance with an embodiment of the present disclosure.

As shown in fig. 3, the method 300 may include the steps of:

s302, providing sample initial voiceprint characteristic data of a preset speaker;

s304, generating a sample initial feature vector of a predetermined speaker based on the sample initial voiceprint feature data;

s306, generating a sample covariance matrix corresponding to the sample initial feature vector;

s308, generating an updated sample feature vector of the preset speaker based on the sample initial feature vector and the sample covariance matrix, wherein the updated sample feature vector is a posterior mean vector of the sample initial feature vector according to Gaussian distribution;

s310, extracting voiceprint characteristics of a preset speaker based on the updated sample characteristic vector; and

s312, network parameters used for updating the voiceprint feature extraction model are obtained based on the voiceprint features, and therefore the voiceprint feature extraction model is trained.

It is noted that the operations of steps S304 to S310 in the method 300 for training the voiceprint feature extraction model are similar to the operations of steps S204 to S210 in the voiceprint feature extraction method 200 described in connection with fig. 2, and are all used for extracting the voiceprint feature of the speaker.

In contrast, in the method 300 for training the voiceprint feature extraction model, the data for training in step S302 is for a predetermined speaker, i.e., sample initial voiceprint feature data, so that the model training can be performed for a specific speaker. Accordingly, speaker tagging needs to be performed on audio data so that it can be determined for which particular speaker the audio data is intended.

In addition, the method 300 for training the voiceprint feature extraction model further includes updating the network parameters of the voiceprint feature extraction model in step S312. In other words, the network parameters are updated by using the voiceprint features extracted in step S310, so that multiple rounds of training can be iterated repeatedly until the network converges, thereby completing the training of the model. Accordingly, the voiceprint features extracted in step S310 can be input to, for example, two fully-connected layers (with ReLU activation functions) and further to the output layer, the loss is calculated by CE (cross entropy) criterion, and finally the network parameters are updated inversely according to SGD (random gradient descent), thus iterating through multiple rounds until the network converges to complete the model training.

FIG. 4 shows a schematic diagram of a voiceprint extraction model 400 in accordance with an embodiment of the present disclosure. The voiceprint feature extraction method and the method for training the voiceprint feature extraction model according to the present disclosure will be further described below with reference to fig. 2 to 4.

As shown in fig. 4, the voiceprint extraction model 400 can include an encoder 410, a gaussian a posteriori inference module 420, and a decoder 430.

The encoder 410 may include a TDNN layer 411, a fully-connected layer 412, and an auxiliary fully-connected layer 413. Fig. 4 schematically shows that the auxiliary full connection layer 413 includes a first auxiliary full connection layer 413-1 and a second auxiliary full connection layer 413-2. However, the number of the auxiliary full connection layers 413 may be appropriately selected according to actual circumstances. The TDNN layer 411 may also be a neural network including multiple layers, which may be selected as appropriate according to the actual situation.

The TDNN layer 411 and the fully-connected layer 412 together may correspond to the first neural network described in connection with fig. 2, and the auxiliary fully-connected layer 413 may correspond to the second neural network described in connection with fig. 2A second neural network as described in connection with fig. 2. As shown in FIG. 4, initial voiceprint feature data X extracted from audio data about a speaker, which may include a plurality of sub-feature data X having a temporal context relationship, may be input to a voiceprint extraction model 400₁,x₂,…x_TThe plurality of sub-feature data x₁,x₂,…x_TMay correspond to T frames after the audio data is framed.

Note that in the training phase of the voiceprint extraction model 400, the initial voiceprint feature data X is sample initial voiceprint feature data corresponding to a particular speaker, since the audio data is speaker-tagged. After the training of the voiceprint extraction model 400 is completed, the audio data may be about any speaker when the voiceprint feature extraction is performed using the voiceprint extraction model 400, and thus the initial voiceprint feature data X is also about any speaker.

As shown in fig. 4, the initial voiceprint feature data X ═ { X ═ X₁,x₂,…,x_TThe initial feature vector z is obtained after passing through the TDNN layer 411 and the fully connected layer 412 (the first neural network)₁,z₂,…,z_T}. Furthermore, the initial feature vector { z }₁,z₂,…,z_TAfter passing through the auxiliary full-connection layer 413, the corresponding initial feature vector z is obtained₁,z₂,…,z_TCovariance matrix log { L }₁,L₂,…,L_T(As mentioned above, the computation of covariance matrix is done logarithmically for ease of computation).

Initial feature vector z₁,z₂,…,z_TEach first element z in₁,z₂,…,z_TFeature information in different feature dimensions may be included. For example, z₁,z₂,…,z_TEach (e.g., in z)₁For example) may contain feature information in a corresponding feature dimension, e.g., related to a physiological characteristic of the speaker's mouth, vocal cords, etc. Thus, the corresponding second element { L } in the covariance matrix₁,L₂,…,L_T} (e.g. correspondingly withL₁For example) may characterize a correlation between vocal features on the mouth and vocal features on the vocal cords, for example.

The operation of the gaussian a posteriori inference module 420 may correspond to generating updated feature vectors for the speaker, shown as φ s in FIG. 4, as described above in connection with FIGS. 2 and 3. In the Gaussian A posteriori inference module 420, the covariance matrix log { L }₁,L₂,…,L_TSoftmax is performed, followed by the initial feature vector z₁,z₂,…,z_TMultiply to obtain the updated feature vector. By covariance matrix log { L }₁,L₂,…,L_TThe Softmax value of { with the initial eigenvector { z }₁,z₂,…,z_TMultiplication is also carried out weighted average operation, so that the obtained updated feature vector phi s is also a posterior mean vector of the initial feature vector according to Gaussian distribution, which is more accurate expression of the voiceprint features of the speaker in Gaussian distribution.

In one example, the prior mean and covariance (μ) may also be used_p，L_p) Initialized to 0 to add to the weighted average calculation as above, i.e. L₀＝0，z₀＝0。

Meanwhile, in the initial voiceprint feature data X ═ { X ═ X₁,x₂,…,x_TIncludes a plurality of sub-feature data { x) having a temporal context relationship₁,x₂,…,x_TIn the case of { z }, the above procedure of obtaining the updated feature vector φ s can represent the initial feature vector { z }₁,z₂,…,z_TA plurality of first elements z having a temporal context in₁,z₂,…,z_TThe polymerization is carried out in time and can therefore also be referred to as time polymerization.

Note that in the training phase of the voiceprint extraction model 400, since the input initial voiceprint feature data X is sample initial voiceprint feature data about a particular speaker, accordingly, the obtained initial feature vector { z }₁,z₂,…,z_TThe covariance matrix log { L }₁,L₂,…,L_TAnd the updated feature vector phis is also the sample initial eigenvector, the sample covariance matrix, and the updated sample eigenvector for the particular speaker, respectively.

Decoder 430 may include an embedding layer 431, a fully-connected layer 432, and an output layer 433. The embedding layer 431 may map the updated feature vector φ s to generate voiceprint features for the speaker.

Note that in the training phase of the voiceprint extraction model 400, a fully connected layer 432 and an output layer 433 may be further included. The voiceprint features extracted at the embedding layer 431 can be input to the fully-connected layer 432 and further to the output layer 433 (for speaker classification), the loss calculated by CE (cross entropy) criterion, and finally the network parameters updated inversely according to SGD (random gradient descent), thus iterating through multiple rounds to network convergence. In the testing phase, voiceprint features may be extracted at the embedding layer 431, and then pairwise similarity may be calculated by cosine (cos).

The voiceprint feature extraction method and the method for training the voiceprint feature extraction model of the present disclosure are described above with reference to fig. 2 to 4. In order that this disclosure may be more fully understood, a brief description of the model assumptions on which the principles of this disclosure are based follows.

First, the assumption of a prior distribution of speaker characteristics is made:

1) model: z is a radical of_t＝h+∈_t

2) Latent variables:

3) uncertainty:

wherein z is_tGenerating variables for the model, representing feature vectors of the speaker; h is a hidden variable representing the basic vector of the speaker; e is the same as_tThe residual variable represents the difference vector of different speakers; μ, L are mean and covariance, respectively, in a Gaussian distribution, and z_tH and e_tObey a gaussian distribution.

The basis vector h mayReflecting the average of the population over several characteristic attributes, such as several characteristic dimensions related to the mouth, vocal cords, tongue, lips, etc. Residual variable ∈_tA perturbation vector that may reflect changes in some of these characteristic properties (e.g., the frequency of vibration of the vocal cords). Based on the model, the characteristics of any speaker can be characterized.

Since the basis vector h obeys a gaussian distribution, the posterior probability of the basis vector h can be calculated for a specific population (e.g., 1000 asians), thereby reflecting the specific shape of the population with respect to the gaussian distribution of the basis vector h. That is, the posterior mean vector of the initial feature vector according to the gaussian distribution is obtained in the present disclosure.

Given input data about a particular population, the posterior probability distribution of basis vector h can be derived as:

wherein:

here, it can be obtained through a neural network

And

wherein f is_enc() And g_enc() May correspond to the first and second neural networks, respectively, as described above.

In addition, the training process of the voiceprint extraction model can also be represented by the following pseudo code:

where θ may represent network parameters to be updated, such as weights and offsets. θ may be calculated by multiplying a parameter g by a learning force, wherein the parameter g may reflect an amount of change or update learned from a batch of data (batch) at a time, and the learning force may represent a magnitude of the change. After the parameter g is obtained, it may be added to θ (e.g., weights and offsets) to produce corresponding changes, thereby implementing the training process.

According to another aspect of the present disclosure, a voiceprint feature extraction apparatus is also provided. Fig. 5 shows a block diagram of a voiceprint feature extraction apparatus 500 according to one embodiment of the present disclosure.

As shown in fig. 5, the apparatus 500 may include: an acquisition unit 502 configured to acquire initial voiceprint feature data about a speaker; a first generating unit 504 configured to generate an initial feature vector of the speaker based on the initial voiceprint feature data; a second generating unit 506 configured to generate a covariance matrix corresponding to the initial eigenvector; a third generating unit 508 configured to generate an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to a gaussian distribution; and an extraction unit 510 configured to extract voiceprint features of the speaker based on the updated feature vector.

According to some embodiments, the initial voiceprint feature data is extracted from audio data about the speaker, and the initial voiceprint feature data includes a plurality of sub-feature data having a temporal context relationship, wherein the plurality of sub-feature data correspond to a plurality of frames of the audio data.

The operations performed by the above-mentioned modules 502 to 510 correspond to the steps S202 to S210 described with reference to fig. 2, and therefore, details of various aspects thereof are not repeated.

Fig. 6 shows a block diagram of a voiceprint feature extraction apparatus 600 according to another embodiment of the present disclosure. The modules 602 to 610 shown in fig. 6 may correspond to the modules 502 to 510 shown in fig. 5, respectively. The modules 604 to 610 may comprise further sub-functional modules in addition thereto, as will be explained in more detail below.

According to some embodiments, the first generating unit 604 may include: a first sub-unit 6040 configured to input the initial voiceprint feature data to a first neural network to obtain an initial feature vector, wherein the initial feature vector includes a plurality of first elements having a temporal context corresponding to the plurality of sub-feature data.

According to some embodiments, the second generating unit 606 may include: a second subunit 6060 configured to input the initial feature vector to a second neural network to obtain a covariance matrix corresponding to the initial feature vector, wherein the covariance matrix has a plurality of second elements corresponding to the plurality of first elements of the initial feature vector, and each second element characterizes a correlation of the corresponding first element between different feature dimensions.

According to some embodiments, the third generating unit 608 may comprise: a third sub-unit 6080 configured to apply a normalized exponent function to the covariance matrix to obtain a normalized exponent value for the covariance matrix; and a fourth subunit 6082 configured to multiply the normalized exponent value with the initial feature vector to obtain an updated feature vector.

According to some embodiments, the extraction unit 610 may include: a mapping unit 6100 configured to map the updated feature vector by an embedding operation to generate voiceprint features of the speaker.

According to another aspect of the present disclosure, there is also provided an apparatus for training a voiceprint feature extraction model. Fig. 7 shows a block diagram of an apparatus 700 for training a voiceprint feature extraction model according to an embodiment of the present disclosure.

As shown in fig. 7, the apparatus 700 may include: a providing unit 702 configured to provide sample initial voiceprint feature data about a predetermined speaker; a first sample generation unit 704 configured to generate a sample initial feature vector of the predetermined speaker based on the sample initial voiceprint feature data; a second sample generation unit 706 configured to generate a sample covariance matrix corresponding to the sample initial feature vector; a third sample generating unit 708 configured to generate an updated sample feature vector of the predetermined speaker based on the sample initial feature vector and the sample covariance matrix, wherein the updated sample feature vector is a posterior mean vector of the sample initial feature vector according to a gaussian distribution; a sample extraction unit 710 configured to extract voiceprint features of a predetermined speaker based on the updated sample feature vector; and a network parameter obtaining unit 712 configured to obtain network parameters for updating the voiceprint feature extraction model based on the voiceprint features to train the voiceprint feature extraction model.

According to another aspect of the present disclosure, there is also provided an electronic device comprising at least one processor; and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor, which when executed by the at least one processor, cause the at least one processor to perform the method as described above.

According to another aspect of the present disclosure, there is also provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method as described above.

According to another aspect of the present disclosure, there is also provided a computer program product comprising a computer program, wherein the computer program realizes the method as described above when executed by a processor.

Referring to fig. 8, a block diagram of a structure of an electronic device 800, which may be a server or a client of the present disclosure, which is an example of a hardware device that may be applied to aspects of the present disclosure, will now be described. Electronic device is intended to represent various forms of digital electronic computer devices, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 8, the apparatus 800 includes a computing unit 801 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)802 or a computer program loaded from a storage unit 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the device 800 can also be stored. The calculation unit 801, the ROM 802, and the RAM 803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

A number of components in the device 800 are connected to the I/O interface 805, including: an input unit 806, an output unit 807, a storage unit 808, and a communication unit 809. The input unit 806 may be any type of device capable of inputting information to the device 800, and the input unit 806 may receive input numeric or character information and generate key signal inputs related to user settings and/or function controls of the electronic device, and may include, but is not limited to, a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and/or a remote control. Output unit 807 can be any type of device capable of presenting information and can include, but is not limited to, a display, speakers, a video/audio output terminal, a vibrator, and/or a printer. The storage unit 808 may include, but is not limited to, a magnetic disk, an optical disk. The communication unit 809 allows the device 800 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers and/or chipsets, such as bluetooth^TMDevices, 1302.11 devices, WiFi devices, WiMax devices, cellular communication devices, and/or the like.

Computing unit 801 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and the like. The calculation unit 801 performs the respective methods and processes described above, such as the voiceprint feature extraction method. For example, in some embodiments, the voiceprint feature extraction method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and/or installed onto device 800 via ROM 802 and/or communications unit 809. When loaded into RAM 803 and executed by computing unit 801, a computer program may perform one or more steps of the voiceprint feature extraction method described above. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the voiceprint feature extraction method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be performed in parallel, sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.

In the technical scheme of the disclosure, the acquisition, storage and application of the personal information of the related user are all in accordance with the regulations of related laws and regulations, and do not violate the good customs of the public order. It is an intention of the present disclosure that personal information data should be managed and processed in a manner that minimizes the risk of inadvertent or unauthorized access to the use. By limiting data collection and deleting data when it is no longer needed, risks are minimized. All information related to a person in the present application is collected with the knowledge and consent of the person.

Although embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it is to be understood that the above-described methods, systems and apparatus are merely exemplary embodiments or examples and that the scope of the present invention is not limited by these embodiments or examples, but only by the claims as issued and their equivalents. Various elements in the embodiments or examples may be omitted or may be replaced with equivalents thereof. Further, the steps may be performed in an order different from that described in the present disclosure. Further, various elements in the embodiments or examples may be combined in various ways. It is important that as technology evolves, many of the elements described herein may be replaced with equivalent elements that appear after the present disclosure.

Claims

1. A voiceprint feature extraction method comprising:

acquiring initial voiceprint characteristic data about a speaker;

generating an initial feature vector of the speaker based on the initial voiceprint feature data;

generating a covariance matrix corresponding to the initial eigenvector;

generating an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to Gaussian distribution; and

extracting voiceprint features of the speaker based on the updated feature vector.

2. The method of claim 1, wherein the initial voiceprint feature data is extracted from audio data about the speaker and the initial voiceprint feature data comprises a plurality of sub-feature data having a temporal context, wherein the plurality of sub-feature data correspond to a plurality of frames of the audio data.

3. The method of claim 2, wherein the generating an initial feature vector for the speaker based on the initial voiceprint feature data comprises:

inputting the initial voiceprint feature data to a first neural network to obtain the initial feature vector, wherein the initial feature vector comprises a plurality of first elements having the temporal context corresponding to the plurality of sub-feature data.

4. The method of claim 3, wherein the generating a covariance matrix corresponding to the initial eigenvector comprises:

inputting the initial feature vector to a second neural network to obtain the covariance matrix corresponding to the initial feature vector, wherein the covariance matrix has a plurality of second elements corresponding to the plurality of first elements of the initial feature vector, and each second element characterizes a correlation between the corresponding first elements in different feature dimensions.

5. The method of any of claims 1-4, wherein the generating the updated feature vector of the speaker comprises:

applying a normalized exponent function to the covariance matrix to obtain a normalized exponent value for the covariance matrix; and

multiplying the normalized index value with the initial feature vector to obtain the updated feature vector.

6. The method of any of claims 1-5, wherein the extracting voiceprint features of the speaker based on the updated feature vector comprises: mapping the updated feature vector by an embedding operation to generate the voiceprint features of the speaker.

7. A method for training a voiceprint feature extraction model, comprising:

providing sample initial voiceprint characteristic data about a predetermined speaker;

generating a sample initial feature vector of the predetermined speaker based on the sample initial voiceprint feature data;

generating a sample covariance matrix corresponding to the sample initial feature vector;

generating an updated sample feature vector of the predetermined speaker based on the sample initial feature vector and the sample covariance matrix, wherein the updated sample feature vector is a posterior mean vector of the sample initial feature vector according to Gaussian distribution;

extracting voiceprint features of the predetermined speaker based on the updated sample feature vector; and

and acquiring network parameters for updating a voiceprint feature extraction model based on the voiceprint features so as to train the voiceprint feature extraction model.

8. A voiceprint feature extraction apparatus comprising:

an acquisition unit configured to acquire initial voiceprint feature data about a speaker;

a first generating unit configured to generate an initial feature vector of the speaker based on the initial voiceprint feature data;

a second generation unit configured to generate a covariance matrix corresponding to the initial feature vector;

a third generating unit configured to generate an updated feature vector of the speaker based on the initial feature vector and the covariance matrix, wherein the updated feature vector is a posterior mean vector of the initial feature vector according to a gaussian distribution; and

an extraction unit configured to extract voiceprint features of the speaker based on the updated feature vector.

9. The apparatus of claim 8, wherein the initial voiceprint feature data is extracted from audio data about the speaker and comprises a plurality of sub-feature data having a temporal context, wherein the plurality of sub-feature data correspond to a plurality of frames of the audio data.

10. The apparatus of claim 9, wherein the first generating unit comprises:

a first subunit configured to input the initial voiceprint feature data to a first neural network to obtain the initial feature vector, wherein the initial feature vector comprises a plurality of first elements having the temporal context corresponding to the plurality of sub-feature data.

11. The apparatus of claim 10, wherein the second generating unit comprises:

a second subunit configured to input the initial feature vector to a second neural network to obtain the covariance matrix corresponding to the initial feature vector, wherein the covariance matrix has a plurality of second elements corresponding to the plurality of first elements of the initial feature vector, and each second element characterizes a correlation of the corresponding first element between different feature dimensions.

12. The apparatus of any of claims 8 to 11, wherein the third generating means comprises:

a third subunit configured to apply a normalized exponent function to the covariance matrix to obtain a normalized exponent value for the covariance matrix; and

a fourth subunit configured to multiply the normalized exponent value with the initial feature vector to obtain the updated feature vector.

13. The apparatus of any one of claims 8 to 12, wherein the extraction unit comprises:

a mapping unit configured to map the updated feature vector by an embedding operation to generate the voiceprint features of the speaker.

14. An apparatus for training a voiceprint feature extraction model, comprising:

a providing unit configured to provide sample initial voiceprint feature data about a predetermined speaker;

a first sample generation unit configured to generate a sample initial feature vector of the predetermined speaker based on the sample initial voiceprint feature data;

a second sample generation unit configured to generate a sample covariance matrix corresponding to the sample initial feature vector;

a third sample generating unit configured to generate an updated sample feature vector of the predetermined speaker based on the sample initial feature vector and the sample covariance matrix, wherein the updated sample feature vector is a posterior mean vector of the sample initial feature vector according to a gaussian distribution;

a sample extraction unit configured to extract voiceprint features of the predetermined speaker based on the updated sample feature vector; and

a network parameter obtaining unit configured to obtain a network parameter for updating a voiceprint feature extraction model based on the voiceprint feature to train the voiceprint feature extraction model.

15. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

16. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-7.

17. A computer program product comprising a computer program, wherein the computer program realizes the method according to any of claims 1-7 when executed by a processor.