Method, device and computer storage medium for secure communication between intercom devices

Abstract

The present application relates to the field of wireless communications, and provides a secure communication method, device, and computer storage medium between intercom devices. The method includes: after establishing a communication connection with an intercom device of the other party, obtaining the call voice data of the user of the intercom device of the other party; using a trained speaker recognition model to extract features of the call voice data to obtain a feature vector of the call voice data; performing a dimensionality reduction operation on the feature vector of the call voice data to obtain an identity feature vector of the user of the intercom device of the other party; based on the identity feature vector of the user of the intercom device of the other party, determining whether the user of the intercom device of the other party is a legitimate user; if the user of the intercom device of the other party is a legitimate user, encrypting the call voice data using the identity feature vector of the user of the intercom device of the other party as a key to obtain encrypted voice data; transmitting the encrypted voice data through the communication connection established with the intercom device of the other party.

Description

Technical Field

[0001] The present application relates to the field of wireless communications, and in particular to a method, device and computer storage medium for secure communication between intercom devices. Background Art

[0002] Two Way Radio is a two-way mobile communication tool that allows calls without network support. It is suitable for relatively fixed and frequent calls. In terms of instant messaging, group communication, real-time and reliability, two-way radio has advantages that mobile terminals such as smartphones do not have. Therefore, two-way radio still has market application prospects in scenarios such as hotels, large venues, schools, supermarkets, buildings, and hospital security.

[0003] Generally speaking, intercom devices communicate with each other through pre-agreed radio frequencies, which are open resources, meaning that third parties may legally or illegally, accidentally or intentionally occupy these open resources. Therefore, the security of communication data between intercom devices, especially those that can easily achieve ultra-long-distance communication of hundreds or even thousands of kilometers, faces great challenges. Summary of the invention

[0004] The present application provides a method, device and computer storage medium for secure communication between intercom devices, which can realize secure communication between intercom devices.

[0005] On the one hand, the present application provides a method for secure communication between intercom devices, the method comprising:

[0006] After establishing a communication connection with the other party's intercom device, obtaining the conversation voice data of the user of the other party's intercom device;

[0007] Using a trained speaker recognition model to extract features from the call voice data to obtain a feature vector of the call voice data;

[0008] Performing a dimensionality reduction operation on the feature vector to obtain an identity feature vector of the user;

[0009] Based on the identity feature vector of the user, determining whether the user is a legitimate user;

[0010] If the user is a legitimate user, encrypting the call voice data using the user's identity feature vector as a key to obtain encrypted voice data;

[0011] The encrypted voice data is re-encrypted using a preset encryption method to obtain re-encrypted voice data;

[0012] The encrypted voice data obtained after the secondary encryption is transmitted through the communication connection established with the intercom device of the other party.

[0013] On the other hand, the present application provides a secure communication device between intercom devices, the device comprising:

[0014] An acquisition module, used to acquire the conversation voice data of the user of the other party's intercom device after establishing a communication connection with the other party's intercom device;

[0015] An extraction module, configured to extract features from the call voice data using a trained speaker recognition model to obtain a feature vector of the call voice data;

[0016] A dimension reduction module, used for performing a dimension reduction operation on the feature vector to obtain an identity feature vector of the user;

[0017] A judgment module, used to judge whether the user is a legitimate user based on the identity feature vector of the user;

[0018] A first encryption module, configured to encrypt the call voice data using the identity feature vector of the user as a key to obtain encrypted voice data if the user is a legitimate user;

[0019] A second encryption module is used to perform secondary encryption on the encrypted voice data using a preset encryption method to obtain secondary encrypted voice data;

[0020] The transmission module is used to transmit the encrypted voice data obtained after the secondary encryption through the communication connection established with the other party's intercom device.

[0021] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the technical solution of the method for secure communication between intercom devices as described above when executing the computer program.

[0022] In a fourth aspect, the present application provides a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of the technical solution of the method for secure communication between intercom devices as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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.

[0024] Figure 1 is a flow chart of a secure communication method between intercom devices provided in an embodiment of the present application;

[0025] Figure 2 It is a structural diagram of a secure communication device between intercom devices provided in an embodiment of the present application;

[0026] Figure 3 It is a schematic diagram of the structure of the device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0027] 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.

[0028] In this specification, adjectives such as first and second may be used only to distinguish one element or action from another element or action, without necessarily requiring or implying any actual such relationship or order. Where circumstances permit, reference to an element or component or step (etc.) should not be interpreted as being limited to only one of the elements, components, or steps, but may be one or more of the elements, components, or steps, etc.

[0029] In this specification, for the convenience of description, the sizes of various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0030] Intercom equipment still has market application prospects in scenarios such as hotels, large conference venues, schools, supermarkets, buildings, and hospital security. Generally speaking, intercom equipment communicates with each other through pre-agreed radio frequencies, and these reserved radio frequencies are open resources, which means that third parties may legally or illegally, accidentally or intentionally occupy these open resources. Therefore, the security of communication data between intercom equipment, especially intercom equipment that can easily achieve ultra-long-distance communication of hundreds or even thousands of kilometers, faces great challenges.

[0031] against Figure 1In order to solve the above problems of the prior art, the present application proposes a method for secure communication between intercom devices, which can realize secure communication between intercom devices. Figure 1 As shown, it mainly includes steps S101 to S107, which are described in detail as follows:

[0032] Step S101: After establishing a communication connection with the other party's intercom device, obtaining the conversation voice data of the user of the other party's intercom device.

[0033] Since the technical solution of the present application is to use any one of the two intercom devices as the execution subject, the other intercom device refers to the intercom device that intends to communicate with the execution subject. The communication connection between the execution subject and the other intercom device can be established by the methods provided by the existing technology, which will not be elaborated here. It should be noted that from the perspective of saving resources and / or improving efficiency, after establishing a communication connection with the other intercom device, the call voice data of the user of the other intercom device is obtained. It is not necessary to obtain all the call voice data of the user of the other intercom device, and it is only necessary to intercept a fragment of the call voice data of the user of the other intercom device.

[0034] Step S102: using the trained speaker recognition model to extract features from the conversation voice data of the user of the other intercom device to obtain a feature vector of the conversation voice data.

[0035] It should be noted that, since the voice signal changes in sequence, the user's call voice data can be divided into signals of a predetermined duration, such as 15 to 25 milliseconds (ms). The signal in each predetermined duration can be regarded as continuous and stable, and each signal of such a predetermined duration is called a frame. The feature extraction of call voice data is to extract the user's voice features and vocal tract characteristics. As an embodiment of the present application, the call voice data of the user of the other party's intercom device is feature extracted using a trained speaker recognition model, and obtaining the feature vector of the call voice data can be achieved through steps S1021 and S1022, as described in detail as follows:

[0036] Step S1021: divide the conversation voice data of the user of the other party's intercom device into frames to obtain at least two frames of conversation voice data.

[0037] As mentioned above, the framing process is to segment the call voice data according to the preset duration and structure it into a stable data structure. The framing of the call voice data can be implemented by using a window function and setting a frame step, etc., which is not limited in this application.

[0038] Step S1022: extracting features from each frame of at least two frames of call voice data according to a predetermined receptive field to obtain a feature vector of the call voice data of the user of the other intercom device.

[0039] In an embodiment of the present application, a partial function network of a trained speaker recognition model can be used to learn features with speaker representation capabilities, thereby directly extracting features from the user's call voice data in an end-to-end manner. In a convolutional neural network, a receptive field refers to the area of ​​the input image that a neuron in a certain layer of the network can see. The size of the receptive field determines the network's ability to perceive local or global information of the input image. In image processing tasks, a smaller receptive field can capture detailed information, while a larger receptive field helps capture a wider range of contextual information. In other words, the larger the receptive field, the more abstract the extracted features. Specifically, step S1022 can be implemented by: using the first sublayer of the partial function network of the trained speaker recognition model to extract features from each frame of call voice data according to the first receptive field to obtain a first vector; using the second sublayer to extract features from each frame of call voice data according to the second receptive field to obtain a second vector; connecting the first vector and the second vector to obtain a feature vector of the call voice data. Here, the second receptive field corresponding to the second sublayer is larger than the first receptive field corresponding to the first sublayer. The final feature vector is a combination of the feature vectors extracted from each sub-layer spliced ​​together according to the weights.

[0040] Step S103: Perform a dimensionality reduction operation on the feature vector of the call voice data to obtain an identity feature vector of the user of the other party's intercom device.

[0041] In an embodiment of the present application, the user's identity feature vector is used to identify the identity of the user of the other party's intercom device. Performing a dimensionality reduction operation on the feature vector of the call voice data to obtain the identity feature vector of the user of the other party's intercom device can be: using a dimensionality reduction network to perform a dimensionality reduction operation on the feature vector of the call voice data to obtain the user's identity feature vector. Specifically, using a dimensionality reduction network to perform a dimensionality reduction operation on the feature vector of the call voice data to obtain the user's identity feature vector can be implemented by: optimizing the feature vector of the call voice data at a frame granularity to obtain a feature vector at a frame granularity; mapping the feature vector at a frame granularity to a feature vector of a preset dimension; connecting the feature vectors of the preset dimension to obtain the user's identity feature vector.

[0042] It should be noted that the trained speaker recognition model that extracts features from the call voice data to obtain the feature vector of the call voice data and performs a dimensionality reduction operation on the feature vector can be obtained by training the speaker recognition model in the following manner: obtaining a speech sample labeled with the speaker identity; obtaining a constructed speaker recognition model, where the constructed speaker recognition model includes an initial feature learning network and an initial dimensionality reduction network; using the model parameters of the trained feature extraction network to initialize the weight parameters of the initial feature learning network; using the speech samples of the speaker identity to train the initial feature learning network and the initial dimensionality reduction network to obtain the speaker recognition model. Among them, the trained feature extraction network is obtained by pre-training the feature extraction network through the following steps: obtaining at least two categories of unlabeled original speech samples; obtaining pseudo-speech samples corresponding to each category of original speech samples; using each category of original speech samples and pseudo-speech samples, self-supervising the initial feature extraction network based on contrast loss; in response to the contrast loss being not greater than a set threshold, determining the model parameters of the feature extraction network to obtain the trained feature extraction network.

[0043] Step S104: Based on the user's identity feature vector, determine whether the user of the other party's intercom device is a legitimate user.

[0044] Specifically, based on the user's identity feature vector, judging whether the user of the other intercom device is a legitimate user can be: matching the identity feature vector of the user of the other intercom device with the identity feature vector of the registered user; if the similarity between the identity feature vector of the user of the other intercom device and the identity feature vector of the registered user exceeds a preset threshold, then determining that the user of the other intercom device is a legitimate user. In the above embodiment, the similarity between the identity feature vector of the user of the other intercom device and the identity feature vector of the registered user can be calculated by a cosine function, that is, the similarity between the identity feature vector of the user of the other intercom device and the identity feature vector of the registered user can be specifically calculated by calculating the cosine similarity between the identity feature vector of the user of the other intercom device and the identity feature vector of the registered user.

[0045] Step S105: If the user of the other party's intercom device is a legitimate user, the call voice data is encrypted using the identity feature vector of the user as a key to obtain encrypted voice data.

[0046] Since the user's identity feature vector is unique, after confirming that the user of the other intercom device is a legitimate user, the identity feature vector of the user of the other intercom device is used as a key to encrypt the call voice data between the other intercom device and the other intercom device. After the encrypted voice data is transmitted, only the user with the identity feature vector of the other intercom device can decrypt the encrypted voice data, while other intercom devices cannot decrypt the encrypted voice data because they do not have the identity feature vector of the user of the other intercom device, thereby ensuring the security of communication between legitimate intercom devices.

[0047] Step S106: The encrypted voice data is re-encrypted using a preset encryption method to obtain re-encrypted voice data.

[0048] Considering that some criminals may obtain the identity feature vector of the user by illegal means, and then use the illegally obtained identity feature vector of the user of the other party's intercom device to decrypt the encrypted voice data, in one embodiment of the present application, after using the identity feature vector of the user of the other party's intercom device as a key to encrypt the call voice data between the other party's intercom device and the other party's intercom device, and obtaining the encrypted voice data, the encrypted voice data can also be re-encrypted using a preset encryption method to obtain the secondary encrypted voice data. As one embodiment of the present application, the encrypted voice data is re-encrypted using a preset encryption method to obtain the secondary encrypted voice data, which can be: obtaining a key corresponding to the user of the other party's intercom device (which has been determined as a legitimate user in step S104) from a preset password table, and using the key corresponding to the legitimate user obtained from the password table to re-encrypt the encrypted voice data to obtain the secondary encrypted voice data. Since criminals cannot or are inconvenient to obtain the preset password table (for example, the password table is regularly updated or randomly changed), the encrypted voice data is re-encrypted using the key corresponding to the user of the other party's intercom device obtained from the password table, so as to prevent criminals from using the illegally obtained identity feature vector of the user of the other party's intercom device to decrypt the encrypted voice data and obtain the call voice data.

[0049] As another embodiment of the present application, the encrypted voice data is re-encrypted using a preset encryption method to obtain the re-encrypted voice data by: obtaining multiple groups of original passwords, and the current group of the multiple groups of original passwords (assuming that the current group is the first group) i Group) original password, according to i The original password of the group i -1 group of original passwords are decrypted, and the decrypted key is used to i-1 group before the original password, and so on, until the first group of original passwords of multiple groups of original passwords are decrypted with the decrypted key to obtain the final encryption password; the encrypted voice data is encrypted with the final encryption password to obtain the secondary encrypted voice data. Since multiple groups of original passwords are used to encrypt the encrypted voice data, and each group of original passwords must be decrypted by the next group of original passwords of the group of original passwords to obtain it, therefore, by adopting a multi-level encryption method for the encryption password of the encrypted voice data, i.e., the final encryption password, the risk of the encrypted voice data or the encryption password of the encrypted voice data being maliciously tampered with or stolen can be reduced, and the security of the encrypted voice data corresponding to the call voice transmission can be improved.

[0050] As another embodiment of the present application, the encrypted voice data is encrypted twice using a preset encryption method, and the obtained twice encrypted voice data may be: according to the volume of the encrypted voice data corresponding to the call voice, a preset noise is added to the encrypted voice data corresponding to the call voice to obtain the twice encrypted voice data. Specifically, according to the volume of the encrypted voice data corresponding to the call voice, a preset noise is added to the encrypted voice data corresponding to the call voice to obtain the twice encrypted voice data: when the volume mean of the encrypted voice data corresponding to the call voice is greater than 0 and not greater than the first volume threshold, the preset regularly changing noise data is mixed into the encrypted voice data to obtain the twice encrypted voice data; when the volume mean of the encrypted voice data corresponding to the call voice is greater than the first volume threshold and not greater than the second volume threshold, the preset transient noise data is mixed into the encrypted voice data to obtain the twice encrypted voice data; when the volume mean of the encrypted voice data corresponding to the call voice is greater than the second volume threshold and not greater than the third volume threshold, the preset slowly changing noise data is mixed into the encrypted voice data to obtain the twice encrypted voice data; when the volume of the encrypted voice data corresponding to the call voice is greater than the third volume threshold, the mitigation noise data is mixed into the encrypted voice data to obtain the twice encrypted voice data. Since regularly changing noise, transient noise, slowly changing noise or moderate noise are all preset or measurable noise data, even if they are added to encrypted voice data, the secondary encrypted voice data received by the legitimate user's receiving end can be decrypted by the legitimate user, but these noises are unknown to attackers and can be kept confidential, thereby improving the security of communication between intercom devices.

[0051] Step S107: Transmitting the secondary encrypted voice data via the communication connection established with the other party's intercom device.

[0052] From the above attached Figure 1It can be seen from the example of a secure communication method between intercom devices that after establishing a communication connection with the other intercom device, the call voice data of the user of the other intercom device is obtained, and then the feature vector of the call voice data is extracted. By performing a dimensionality reduction operation on the feature vector of the call voice data, the user's identity feature vector is obtained, and then based on the user's identity feature vector, it is determined whether the user is a legitimate user. Since the user's identity feature vector is unique, the user's identity feature vector is used as a key to encrypt the call voice between the intercom device and the other party. Even if the encrypted voice data is intercepted during the transmission process, the intercepting party cannot decrypt it. Therefore, the technical solution provided in the embodiment of the present application can ensure the security of communication between intercom devices, especially intercom devices for ultra-long-distance communication.

[0053] Please refer to the attached Figure 2 , is a secure communication device between intercom devices provided in an embodiment of the present application, the device may include an acquisition module 201, an extraction module 202, a dimensionality reduction module 203, a judgment module 204, a first encryption module 205, a second encryption module 206 and a transmission module 207, which are described in detail as follows:

[0054] An acquisition module 201 is used to acquire the conversation voice data of the user of the other party's intercom device after establishing a communication connection with the other party's intercom device;

[0055] An extraction module 202, configured to extract features from the call voice data using a trained speaker recognition model to obtain a feature vector of the call voice data;

[0056] Dimensionality reduction module 203, used to perform dimensionality reduction operation on the feature vector of the call voice data to obtain the identity feature vector of the user of the other party's intercom device;

[0057] A judgment module 204, configured to judge whether the user of the other party's intercom device is a legitimate user based on the identity feature vector of the user of the other party's intercom device;

[0058] The first encryption module 205 is used to encrypt the call voice data using the identity feature vector of the other party's intercom device as a key to obtain encrypted voice data if the user of the other party's intercom device is a legitimate user;

[0059] The second encryption module 206 is used to perform secondary encryption on the encrypted voice data using a preset encryption method to obtain secondary encrypted voice data;

[0060] The transmission module 207 is used to transmit the secondary encrypted voice data through the communication connection established with the other party's intercom device.

[0061] From the above attached Figure 2It can be seen from the example of a secure communication device between intercom devices that after establishing a communication connection with the other intercom device, the call voice data of the user of the other intercom device is obtained, and then the feature vector of the call voice data is extracted. By performing a dimensionality reduction operation on the feature vector of the call voice data, the user's identity feature vector is obtained, and then based on the user's identity feature vector, it is determined whether the user is a legitimate user. Since the user's identity feature vector is unique, the user's identity feature vector is used as a key to encrypt the call voice between the intercom device and the other party. Even if the encrypted voice data is intercepted during the transmission process, the intercepting party cannot decrypt it. Therefore, the technical solution provided in the embodiment of the present application can ensure the security of communication between intercom devices, especially intercom devices for ultra-long-distance communication.

[0062] Optionally, the above Figure 2 In the example device, the trained speaker recognition model is trained through the following steps: obtaining speech samples labeled with speaker identities; obtaining a constructed speaker recognition model, wherein the constructed speaker recognition model includes an initial feature learning network and an initial dimensionality reduction network; using the model parameters of the trained feature extraction network to initialize the weight parameters of the initial feature learning network; using speech samples with speaker identities to train the initial feature learning network and the initial dimensionality reduction network to obtain the speaker recognition model.

[0063] Optionally, the second encryption module may include a key acquisition unit and a preset key encryption unit, wherein:

[0064] A key acquisition unit, used to acquire a key corresponding to a legitimate user from a preset password table;

[0065] The preset key encryption module is used to perform secondary encryption on the encrypted voice data using a key corresponding to a legitimate user obtained from a password table.

[0066] Optionally, the second encryption module may include a multi-stage key acquisition unit and a final key encryption unit, wherein:

[0067] The multi-level key acquisition unit is used to obtain multiple groups of original passwords, and the current group of the multiple groups of original passwords (assuming that the current group is the first i Group) original password, according to i The original password of the group i -1 group of original passwords are decrypted, and the decrypted key is used to i -1 group of original passwords before it is decrypted, and so on, until the first group of original passwords of multiple groups of original passwords is decrypted with the decrypted key to obtain the final encrypted password;

[0068] The final key encryption unit is used to encrypt the encrypted voice data using the final encryption password to obtain secondary encrypted voice data.

[0069] Optionally, the second encryption module may include a noise encryption unit for adding preset noise to the call voice corresponding to the encrypted voice data according to the volume of the call voice corresponding to the encrypted voice data to obtain secondary encrypted voice data.

[0070] Figure 3 Schematic diagram of the structure of the device provided by an embodiment of the present application. Figure 3 As shown, the device 3 of this embodiment mainly includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30, such as a program for a secure communication method between intercom devices. When the processor 30 executes the computer program 32, the steps in the embodiment of the secure communication method between intercom devices are implemented, such as Figure 1 Alternatively, when the processor 30 executes the computer program 32, the functions of each module / unit in the above-mentioned device embodiments are implemented, for example Figure 2 The functions of the acquisition module 201, the extraction module 202, the dimensionality reduction module 203, the judgment module 204, the first encryption module 205, the second encryption module 206 and the transmission module 207 are shown.

[0071] Exemplarily, the computer program 32 of the secure communication method between intercom devices mainly includes: after establishing a communication connection with the other intercom device, obtaining the call voice data of the user of the other intercom device; using the trained speaker recognition model to extract features of the call voice data to obtain the feature vector of the call voice data; performing a dimensionality reduction operation on the feature vector of the call voice data to obtain the identity feature vector of the user of the other intercom device; judging whether the user of the other intercom device is a legitimate user based on the identity feature vector of the user of the other intercom device; if the user of the other intercom device is a legitimate user, using the identity feature vector of the user as a key to encrypt the call voice data to obtain encrypted voice data; using a preset encryption method to perform secondary encryption on the encrypted voice data to obtain secondary encrypted voice data; transmitting the secondary encrypted voice data through the communication connection established with the other intercom device. The computer program 32 can be divided into one or more modules / units, one or more modules / units are stored in the memory 31, and are executed by the processor 30 to complete the present application. One or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program 32 in the device 3. For example, the computer program 32 can be divided into the functions of an acquisition module 201, an extraction module 202, a dimensionality reduction module 203, a judgment module 204, a first encryption module 205, a second encryption module 206 and a transmission module 207 (modules in the virtual device), and the specific functions of each module are as follows: the acquisition module 201 is used to acquire the call voice data of the user of the other party's intercom device after establishing a communication connection with the other party's intercom device; the extraction module 202 is used to use the trained speaker recognition model to extract features from the call voice data to obtain the feature vector of the call voice data; the dimensionality reduction module 203 is used to reduce the feature vector of the call voice data The first encryption module 205 is used to encrypt the call voice data with the identity feature vector of the user of the other party's intercom device as a key to obtain encrypted voice data if the user of the other party's intercom device is a legal user; the second encryption module 206 is used to perform secondary encryption on the encrypted voice data using a preset encryption method to obtain secondary encrypted voice data; and the transmission module 207 is used to transmit the secondary encrypted voice data through the communication connection established with the other party's intercom device.

[0072] The device 3 may include but is not limited to a processor 30 and a memory 31. Those skilled in the art will appreciate that Figure 3It is only an example of device 3 and does not constitute a limitation of device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, a computing device may also include input and output devices, network access devices, buses, etc.

[0073] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0074] The memory 31 may be an internal storage unit of the device 3, such as a hard disk or memory of the device 3. The memory 31 may also be an external storage device of the device 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the device 3. Further, the memory 31 may also include both an internal storage unit and an external storage device of the device 3. The memory 31 is used to store computer programs and other programs and data required by the device. The memory 31 may also be used to temporarily store data that has been output or is to be output.

[0075] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0076] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0077] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0078] In the embodiments provided in the present application, it should be understood that the disclosed devices / equipment and methods can be implemented in other ways. For example, the device / equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0079] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0080] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0081] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing related hardware through a computer program. The computer program of the secure communication method between intercom devices can be stored in a computer storage medium. When the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented, that is, after establishing a communication connection with the other intercom device, obtaining the call voice data of the user of the other intercom device; using the trained speaker recognition model to extract features of the call voice data to obtain the feature vector of the call voice data; performing a dimensionality reduction operation on the feature vector of the call voice data to obtain the identity feature vector of the user of the other intercom device; based on the identity feature vector of the user of the other intercom device, judging whether the user of the other intercom device is a legitimate user; if the user of the other intercom device is a legitimate user, using the identity feature vector of the user as a key to encrypt the call voice data to obtain encrypted voice data; using a preset encryption method to re-encrypt the encrypted voice data to obtain re-encrypted voice data; transmitting the re-encrypted voice data through the communication connection established with the other intercom device. Among them, computer programs include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. Computer storage media may include: any entity or device capable of carrying computer program codes, recording media, USB flash drives, mobile hard disks, magnetic disks, optical disks, computer memories, read-only memories (ROM), random access memories (RAM), electric carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the contents of computer storage media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer storage media do not include electric carrier signals and telecommunication signals.

[0082] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application is described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. The specific implementation methods described above further describe the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application, and is not used to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present invention.

Claims

1. A secure communication method between intercom devices, characterized in that: The method comprises: After establishing a communication connection with the other party's intercom device, intercepting a segment of the conversation voice data of the user of the other party's intercom device; Using the trained speaker recognition model to extract features from the segment of the call voice data to obtain a feature vector of the segment of the call voice data; Performing a dimensionality reduction operation on the feature vector to obtain an identity feature vector of the user; Based on the identity feature vector of the user, determining whether the user is a legitimate user; If the user is a legitimate user, the call voice data is encrypted using the identity feature vector of the user as a key to obtain encrypted voice data; The encrypted voice data is re-encrypted using a preset encryption method to obtain re-encrypted voice data; The secondary encrypted voice data is transmitted through the communication connection established with the other party's intercom device.

2. The secure communication method between intercom devices as claimed in claim 1, characterized in that: The step of extracting features from a segment of the call voice data using a trained speaker recognition model to obtain a feature vector of the call voice data includes: Divide the segments of the call voice data into frames to obtain at least two frames of call voice data; Feature extraction is performed on each frame of call voice data in the segment of the at least two frames of call voice data according to a predetermined receptive field to obtain a feature vector of the segment of the call voice data.

3. The secure communication method between intercom devices as claimed in claim 2, characterized in that: The performing a dimensionality reduction operation on the feature vector to obtain the identity feature vector of the user includes: performing a dimensionality reduction operation on the feature vector of the segment of the call voice data using a dimensionality reduction network to obtain the identity feature vector of the user.

4. The secure communication method between intercom devices as claimed in claim 3, characterized in that: The step of using a dimensionality reduction network to perform a dimensionality reduction operation on the feature vector of the segment of the call voice data to obtain the identity feature vector of the user includes: Performing frame granularity optimization processing on the feature vector of the segment of the call voice data to obtain a frame granularity feature vector; Mapping the feature vector of the frame granularity to a feature vector of a preset dimension; The feature vectors of the preset dimensions are connected to obtain the identity feature vector of the user.

5. The secure communication method between intercom devices as claimed in claim 1, characterized in that: The determining whether the user is a legitimate user based on the identity feature vector of the user includes: Matching the user's identity feature vector with registered user identity feature vectors; If the similarity between the identity feature vector of the user and the identity feature vector of the registered user exceeds a preset threshold, the user is determined to be a legitimate user.

6. The secure communication method between intercom devices as claimed in claim 1, characterized in that: The trained speaker recognition model is trained by the following steps: Obtain speech samples labeled with the speaker's identity; Acquire a constructed speaker recognition model, wherein the constructed speaker recognition model includes an initial feature learning network and an initial dimensionality reduction network; Initializing the weight parameters of the initial feature learning network using the model parameters of the trained feature extraction network; The initial feature learning network and the initial dimension reduction network are trained using the speech samples to obtain the speaker recognition model.

7. The secure communication method between intercom devices as claimed in claim 1, characterized in that: The method of performing secondary encryption on the encrypted voice data by using a preset encryption method to obtain secondary encrypted voice data includes: Obtaining a key corresponding to the legal user from a preset password table; The encrypted voice data is re-encrypted using a key corresponding to the legitimate user to obtain re-encrypted voice data.

8. A secure communication device between intercom devices, characterized in that: The device comprises: An acquisition module, used to intercept a segment of the conversation voice data of the user of the other party's intercom device after establishing a communication connection with the other party's intercom device; An extraction module, configured to extract features from the segments of the call voice data using a trained speaker recognition model to obtain feature vectors of the segments of the call voice data; A dimension reduction module, used for performing a dimension reduction operation on the feature vector to obtain an identity feature vector of the user; A judgment module, used to judge whether the user is a legitimate user based on the identity feature vector of the user; A first encryption module, configured to encrypt the call voice data using the identity feature vector of the user as a key to obtain encrypted voice data if the user is a legitimate user; A second encryption module is used to perform secondary encryption on the encrypted voice data using a preset encryption method to obtain secondary encrypted voice data; The transmission module is used to transmit the secondary encrypted voice data through the communication connection established with the other party's intercom device.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images