Voiceprint verification method and device, electronic equipment and readable storage medium

Abstract

The invention discloses a voiceprint verification method and device, electronic equipment and a readable storage medium, and belongs to the technical field of electronic equipment. The method comprises the following steps: collecting a voice signal for voiceprint verification, and obtaining voiceprint feature information of the voice signal; under the condition that the voiceprint feature information is matched with the preset voiceprint feature information, detecting a low-frequency signal of the spray microphone on the voice signal; and determining that the voiceprint verification of the voice signal is successful under the condition that the voice signal contains the spray microphone low-frequency signal.

Description

Technical Field

[0001] The present application belongs to the technical field of electronic equipment, and specifically relates to a voiceprint verification method, device, electronic equipment and readable storage medium. Background Art

[0002] Voiceprint recognition technology is a technology that extracts the characteristics of human voice signals to identify the speaker. At present, voiceprint recognition technology has been widely used in identity verification, voice assistants, call noise reduction, meeting records and other fields.

[0003] However, due to the particularity of voice, voiceprint verification based on voiceprint recognition technology has the risk of being easily misused compared to other forms of identity verification. For example, a voice with a high degree of similarity to the user's voice can be obtained through recording or voiceprint replication technology to successfully pass the voiceprint verification, resulting in poor security of voiceprint verification. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a voiceprint verification method, device, electronic device and readable storage medium, which can improve the security of voiceprint verification.

[0005] In a first aspect, an embodiment of the present application provides a voiceprint verification method, the method comprising: collecting a voice signal for voiceprint verification, and obtaining voiceprint feature information of the voice signal; when the voiceprint feature information matches preset voiceprint feature information, performing a detection of a microphone low-frequency signal on the voice signal; when the voice signal contains a microphone low-frequency signal, determining that the voiceprint verification of the voice signal is successful.

[0006] In the second aspect, an embodiment of the present application provides a voiceprint verification device, which includes an acquisition module, a detection module and a determination module; the acquisition module is used to collect a voice signal for voiceprint verification and obtain voiceprint feature information of the voice signal; the detection module is used to perform a detection of a microphone low-frequency signal on the voice signal when the voiceprint feature information matches the preset voiceprint feature information; the determination module is used to determine that the voiceprint verification of the voice signal is successful when the voice signal contains a microphone low-frequency signal.

[0007] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.

[0008] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0009] In a fifth aspect, an embodiment of the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect.

[0010] In a sixth aspect, an embodiment of the present application provides a computer program / program product, which is stored in a storage medium and is executed by at least one processor to implement the method described in the first aspect.

[0011] In the embodiment of the present application, a voice signal for voiceprint verification can be collected, and the voiceprint feature information of the voice signal can be obtained; and when the voiceprint feature information matches the preset voiceprint feature information, the voice signal is detected for the low-frequency signal of the microphone; and when the voice signal contains the low-frequency signal of the microphone, it is determined that the voiceprint verification of the voice signal is successful. Through this scheme, since when the voiceprint feature information of the voice signal used for voiceprint verification matches the preset voiceprint feature information, it can be detected whether the voice signal contains the low-frequency signal of the microphone, and the low-frequency signal of the microphone is usually generated by a real person, so when the voice signal contains the low-frequency signal of the microphone, it is determined that the voiceprint verification of the voice signal is successful. It can be determined that the voiceprint verification can be successfully passed on the premise of ensuring that the voice signal is a voice signal generated by a real person rather than a recording or a replica sound, thereby avoiding the impersonation of a recording or a replica sound, which can improve the security of voiceprint verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is one of the flow charts of the voiceprint verification method provided in the embodiment of the present application;

[0013] Figure 2 It is a spectrogram of a speech signal collected when the user's mouth is about 5 cm away from the microphone in the voiceprint verification method provided in an embodiment of the present application;

[0014] Figure 3 It is an enlarged schematic diagram of the spectrum of the low-frequency signal of the microphone in the voiceprint verification method provided in the embodiment of the present application;

[0015] Figure 4 It is a spectrogram of a user voice signal played by a speaker collected in the voiceprint verification method provided in an embodiment of the present application;

[0016] Figure 5 This is the second flow chart of the voiceprint verification method provided in the embodiment of the present application;

[0017] Figure 6 This is the third flow chart of the voiceprint verification method provided in the embodiment of the present application;

[0018] Figure 7 This is one of the schematic diagrams of a method for performing voice activation detection on a voice signal in the voiceprint verification method provided in an embodiment of the present application;

[0019] Figure 8 This is the second schematic diagram of a method for performing voice activation detection on a voice signal in the voiceprint verification method provided in an embodiment of the present application;

[0020] Fig. 9 This is the fourth flow chart of the voiceprint verification method provided in the embodiment of the present application;

[0021] Fig.10 It is a schematic diagram of the correspondence between a voice and voice intensity information corresponding to the voice in the voiceprint verification method provided in an embodiment of the present application;

[0022] Fig.11 This is the fifth flow chart of the voiceprint verification method provided in the embodiment of the present application;

[0023] Fig.12 This is the sixth flow chart of the voiceprint verification method provided in the embodiment of the present application;

[0024] Fig.13 is a schematic diagram of a voiceprint verification device provided in an embodiment of the present application;

[0025] Fig.14 is a schematic diagram of an electronic device provided in an embodiment of the present application;

[0026] Fig.15 It is a hardware schematic diagram of the electronic 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 describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0028] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0029] The term "indication" in this application can be a direct indication (or explicit indication) or an indirect indication (or implicit indication). A direct indication can be understood as the sender explicitly informing the receiver of specific information, operations to be performed, or request results in the sent indication; an indirect indication can be understood as the receiver determining the corresponding information according to the indication sent by the sender, or making a judgment and determining the operation to be performed or the request result according to the judgment result.

[0030] The terms "at least one (item)", "at least one of" and the like in the specification and claims of the present application refer to any one, any two or a combination of more than two of the objects included therein. For example, at least one (item) of a, b, and c can be represented by: "a", "b", "c", "a and b", "a and c", "b and c" and "a, b and c", where a, b, and c can be single or multiple. Similarly, "at least two (items)" refers to two or more, and its meaning is similar to that of "at least one (item)".

[0031] The voiceprint verification method, device, electronic device and readable storage medium provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings through specific embodiments and their application scenarios.

[0032] Voiceprint recognition is a technology that extracts the characteristics of human voice signals to identify the speaker. It only requires microphone equipment and is low-cost, making it an economical and convenient identity recognition technology.

[0033] At present, voiceprint recognition technology has been widely used in identity authentication, voice assistant, call noise reduction, meeting records and other fields. In these scenarios, voiceprint recognition technology can provide convenient and efficient identity authentication and control means, and has broad application prospects in many fields.

[0034] Although voiceprint recognition technology has many advantages such as convenience, high efficiency, and low cost, due to the particularity of voice, voiceprint verification based on voiceprint recognition technology is more vulnerable to fraud than other forms of identity verification. For example, voiceprints can be imitated by recording, thereby deceiving voiceprint verification. With the continuous development of artificial intelligence technology, voiceprint replication technology has also made great breakthroughs. Through a few short sentences from the user, the artificial intelligence model can simulate the user's voice with a high degree of similarity to successfully pass the voiceprint verification. This results in poor security of voiceprint verification.

[0035] In order to solve the above problems, the embodiments of the present application provide a voiceprint verification method, device, electronic device and readable storage medium. The voiceprint verification method provided in the embodiments of the present application can be applied in the scenario of unlocking the screen of an electronic device through voiceprint verification.

[0036] For example, a voice signal A for voiceprint verification can be collected first, and the voiceprint feature information of the voice signal A can be obtained; when the voiceprint feature information matches the preset voiceprint feature information, the voice signal A can be tested for a low-frequency signal; and when the voice signal A contains a low-frequency signal, it is determined that the voiceprint verification of the voice signal A is successful, so that the screen of the electronic device can be successfully unlocked. Through this solution, since when the voiceprint feature information of the voice signal for voiceprint verification matches the preset voiceprint feature information, it can be detected first whether the voice signal contains a low-frequency signal, and the low-frequency signal is usually generated by a real person, so when the voice signal contains a low-frequency signal, it is determined that the voiceprint verification of the voice signal is successful. It can be determined that the voiceprint verification can be successfully passed on the premise of ensuring that the voice signal is a voice signal generated by a real person rather than a recording or a replica sound, thereby avoiding the impersonation of a recording or a replica sound, thereby improving the security of voiceprint verification and protecting user privacy information.

[0037] Furthermore, the voiceprint verification method provided in the embodiment of the present application may not rely on the upgrade of hardware and voiceprint algorithms, thereby improving the security of voiceprint verification without increasing hardware costs and algorithm complexity.

[0038] It should be noted that the voiceprint verification method provided in the embodiments of the present application can be executed by a voiceprint verification device, an electronic device, or a functional module in an electronic device, etc. In some embodiments of the present application, an electronic device executing the voiceprint verification method is taken as an example to illustrate the voiceprint verification method provided in the embodiments of the present application.

[0039] Figure 1 The flowchart of the voiceprint verification method provided by the embodiment of the present application is shown. Figure 1 As shown, the voiceprint verification method provided in the embodiment of the present application may include the following steps 101 to 103.

[0040] Step 101: The electronic device collects a voice signal for voiceprint verification and obtains voiceprint feature information of the voice signal.

[0041] Optionally, in an embodiment of the present application, the above-mentioned voiceprint verification can be used for any identity verification operation such as unlocking the screen of an electronic device, payment operations, etc.

[0042] For example, if the voiceprint verification is successful, the screen of the electronic device can be unlocked; if the voiceprint verification fails, the screen of the electronic device cannot be unlocked.

[0043] Optionally, in the embodiment of the present application, the above-mentioned voice signal may be a voice signal used to pass the above-mentioned voiceprint verification.

[0044] For example, the voice signal may be a voice signal of “I am the owner of the device”. If the voiceprint verification of the voice signal is successful, relevant operations corresponding to the voiceprint verification may be performed.

[0045] Optionally, in an embodiment of the present application, when the electronic device collects the above-mentioned voice signal, the user can place the microphone of the electronic device about 5 cm in front of the mouth, and then speak the voice corresponding to the voice signal to ensure that there is enough breath to be detected by the microphone.

[0046] Optionally, in an embodiment of the present application, after collecting the above-mentioned voice signal, the electronic device can input the voice signal into a voiceprint verification algorithm, and the voiceprint verification algorithm can extract voiceprint feature information of the voice signal.

[0047] It should be noted that the above-mentioned voiceprint verification algorithm is used to perform voiceprint verification on speech signals, and mainly includes three parts: feature extraction, model construction and score calculation. For the specific description of the voiceprint verification algorithm, please refer to the relevant description in the relevant technology. In order to avoid repetition, it will not be repeated here.

[0048] Optionally, in the embodiment of the present application, the above-mentioned voiceprint feature information may include but is not limited to at least one of the following: spectrum, rhythm, linguistic features, etc.

[0049] Step 102: When the voiceprint feature information matches the preset voiceprint feature information, the electronic device performs microphone low-frequency signal detection on the voice signal.

[0050] Optionally, in an embodiment of the present application, the detection of the above-mentioned microphone-spraying low-frequency signal is used to detect whether the voice signal contains the microphone-spraying low-frequency signal.

[0051] Optionally, in an embodiment of the present application, the above-mentioned microphone low-frequency signal can be used to characterize whether the speech signal is a speech signal uttered by a real person.

[0052] It should be noted that if the user's mouth is close to the microphone when speaking, the airflow exhaled by the user will impact the microphone, forming a unique microphone splashing sound. Figure 2 As shown in FIG. 1 , from the sound spectrum, the microphone spraying during speaking will form a unique microphone spraying low-frequency signal 21 in the low-frequency region of 0-100 Hz; Figure 3 As shown, the low-frequency signal 31 is Figure 2 The medium spraying low frequency signal 21 is the spraying low frequency signal after amplification; and Figure 4 As shown, these low-frequency signals cannot be generated by playing through loudspeakers. By judging whether these low-frequency signals exist in the user's voice signal, it is possible to distinguish whether the voice is a real person's voice or a loudspeaker simulating the user's voice.

[0053] Optionally, in an embodiment of the present application, if the above-mentioned voice signal includes the above-mentioned low-frequency signal of the microphone, then the voice signal can be considered to be a voice signal of a real person; if the above-mentioned voice signal does not include the above-mentioned low-frequency signal of the microphone, then the voice signal can be considered not to be a voice signal of a real person, and may be a recorded or AI-generated voice signal.

[0054] Optionally, in an embodiment of the present application, the preset voiceprint feature information may be: during the voiceprint pre-training of the voiceprint verification, a voice signal input by the user is collected, and the voiceprint feature information of the voice signal is extracted.

[0055] Optionally, in an embodiment of the present application, the above-mentioned voiceprint feature information matches the above-mentioned preset voiceprint feature information, which can be understood as: the matching degree or similarity between the voiceprint feature information and the preset voiceprint feature information is greater than or equal to a first threshold; wherein, the first threshold can be preset by the system, or can be arbitrarily set by the user according to actual usage requirements.

[0056] Optionally, in an embodiment of the present application, the electronic device may compare the voiceprint feature information with the preset voiceprint feature information through the voiceprint verification algorithm. If the degree of match or similarity between the voiceprint feature information and the preset voiceprint feature information is greater than or equal to the first threshold, it can be determined that the voiceprint feature information matches the preset voiceprint feature information.

[0057] It can be understood that when the above voiceprint feature information matches the above preset voiceprint feature information, it can be considered that the voiceprint feature information is consistent with the voiceprint feature information of the user himself.

[0058] Optionally, in an embodiment of the present application, when the voiceprint feature information does not match the preset voiceprint feature information, it can be considered that the voiceprint feature information is inconsistent with the user's own voiceprint feature information. At this time, the electronic device can determine that the voiceprint verification of the voice signal has failed.

[0059] Optionally, in the embodiment of the present application, the voiceprint feature information does not match the preset voiceprint feature information, which can be understood as: the matching degree or similarity between the voiceprint feature information and the preset voiceprint feature information is less than the first threshold.

[0060] Optionally, in the embodiment of the present application, combined with Figure 1 ,like Figure 5 As shown, the above step 102 can be specifically implemented through the following step 102a.

[0061] Step 102a: When the voiceprint feature information matches the preset voiceprint feature information, the electronic device performs microphone low-frequency signal detection on the speech signal through the breath detection model.

[0062] Among them, the above-mentioned breath detection model (also known as breath detection algorithm) is a model obtained by training a Gaussian mixture model or a deep learning model using at least two microphone speech signals.

[0063] Optionally, in an embodiment of the present application, the above-mentioned deep learning model can be a convolutional neural network or a recurrent neural network, etc.

[0064] Optionally, in an embodiment of the present application, the above-mentioned breath detection model may be a model that can realize a breath detection function (ie, detect whether a speech signal contains a low-frequency signal of a microphone).

[0065] It should be noted that the embodiments of the present application are merely illustrative of the example of realizing the breath detection function through the above-mentioned breath detection model. In actual implementation, the breath detection function can also be realized through any other possible way, and the embodiments of the present application are not limited thereto.

[0066] Exemplarily, the implementation of the above breath detection model mainly includes the following steps:

[0067] Step 1: Collect and annotate a large amount of microphone speech data in advance, and extract features from the 0-100Hz spectrum signal in these microphone speech data. Common feature extraction methods include Short-Time Fourier Transform (STFT) or Mel Frequency Cepstrum Coefficient (MFCC).

[0068] Step 2: Feature preprocessing: After extracting features, these features usually need to be preprocessed, such as normalization, noise reduction, smoothing, etc., in order to convert the feature data into a form suitable for classification.

[0069] Step 3. Select a classifier: Select a suitable classifier to classify the features. In speech signal processing, commonly used classifiers include Gaussian Mixture Module (GGM) or deep learning models, which may include Convolutional Neural Networks (CNN), Recurrent Neural Network (RNN), etc. The classifier will extract features from the original speech data and convert them into representative feature vectors.

[0070] Step 4, model training and evaluation: Taking the CNN deep learning model as an example, using the labeled speech data (including the speech data of the microphone low-frequency signal), the model is constructed through the model (including the neural network structure, number of layers, etc.), and the model is initialized. At this stage, the classifier will learn the relationship between categories based on the labeled training data set, and then use another set of unused data to evaluate and verify the model, and adjust the model parameters to make it better fit the data.

[0071] Step 5, model optimization: In this stage, the model is tested using actual speech input. By minimizing the loss function, the model can learn to adjust its parameters and reduce the difference between the model prediction and the actual observation, i.e., the loss, to improve the accuracy of the prediction and continuously optimize the model. When the model false rejection rate (FRR) drops below a certain value (e.g., 5%) and remains stable, and the false acceptance rate (FAR) drops below a certain value (e.g., 3%) and remains stable, the model can be considered to meet the usage requirements.

[0072] After completing the above training, a breath detection model for identifying low-frequency signals of microphone spraying can be obtained. The input and output of the breath detection model are as follows:

[0073] Input: a continuous audio signal;

[0074] Output: whether the audio signal contains the low-frequency signal generated by the voice microphone;

[0075] After the new input data is feature extracted, the classifier will assign it to the corresponding category or label based on the learned model parameters. In the subsequent breath detection process, the breath detection model can be used to detect whether there is a low-frequency signal in the voice data.

[0076] In the embodiment of the present application, since the electronic device can detect whether a speech signal contains a low-frequency signal through the above-mentioned breath detection model, it is only necessary to input the speech signal into the breath detection model to directly detect whether the speech signal contains a low-frequency signal, thereby simplifying the process and operation steps of detecting whether a speech signal contains a low-frequency signal; and the breath detection model is obtained by training a large amount of data, thereby improving the accuracy of detecting whether a speech signal contains a low-frequency signal.

[0077] Step 103: When the voice signal includes a low-frequency signal, the electronic device determines that the voiceprint verification of the voice signal is successful.

[0078] Optionally, in an embodiment of the present application, when the above-mentioned voice signal includes the above-mentioned microphone low-frequency signal, the voice signal can be considered to be a voice signal uttered by a real person, and the voice signal matches the above-mentioned preset voice signal, so that the electronic device can determine that the voiceprint verification of the voice signal is successful, and then the electronic device can perform subsequent operations such as unlocking the screen of the electronic device or payment operations after the voiceprint verification is successful.

[0079] In this way, without relying on hardware and voiceprint algorithm upgrades, by detecting whether the user's voice contains low-frequency signals of the microphone, it is possible to effectively distinguish whether the voice is made by a real person, thereby reducing the risk of voiceprint impersonation in the voiceprint verification process, thereby improving the security of voiceprint verification, enhancing user experience, and protecting user privacy.

[0080] Optionally, in the embodiment of the present application, combined with Figure 1 ,like Figure 6 As shown, the above step 103 can be specifically implemented through the following step 103a.

[0081] Step 103a: When the voice signal includes a low-frequency signal and the time domain range of the low-frequency signal is within the time domain range of the voice signal, the electronic device determines that the voiceprint verification of the voice signal is successful.

[0082] Optionally, in an embodiment of the present application, when there are multiple low-frequency signals for spraying the microphone, the time domain range of the low-frequency signals for spraying the microphone is within the time domain range of the voice signal. It can be understood that the time domain range of each low-frequency signal for spraying the microphone is within the time domain range of the voice signal.

[0083] Optionally, in an embodiment of the present application, after determining that the voice signal contains the microphone low-frequency signal, the electronic device may perform voice activity detection (VAD) on the voice signal to determine that the time domain range of the microphone low-frequency signal is within the time domain range of the voice signal.

[0084] For example, Figure 7As shown, through the above VAD, the front end point V1 and the rear end point V2 of the above voice signal can be obtained. In the process of breath detection, in addition to detecting whether the low-frequency signal of the microphone is present, it is also necessary to output the starting point S1 and the ending point E1 of the low-frequency signal of the microphone. Since there may be multiple groups of such breath sounds in the voice signal, multiple groups of data such as S1 and E1, S2 and E2 are obtained. After obtaining these data, the starting point and the ending point of the multiple groups of low-frequency signals of the microphone can be compared with the front end point V1 and the rear end point V2 of the voice signal to determine whether the time domain of all low-frequency signals of the microphone are included in the front end point V1 and the rear end point V2 of the voice signal. The time domain of all low-frequency signals of the microphone are included in the front end point V1 and the rear end point V2 of the voice signal. At this time, it is considered that the low-frequency signal of the microphone in this verification is strongly coupled with the voice signal, and the low-frequency signal of the microphone is generated by the user's voice, and the voiceprint verification passes; otherwise, as Figure 8 As shown, the time domain of all the low-frequency signals of the microphone are not included by the front endpoint V1 and the rear endpoint V2 of the voice signal. At this time, it is considered that the low-frequency signals of the microphone are generated by an irrelevant environment and the voiceprint verification fails.

[0085] Optionally, in an embodiment of the present application, if the time domain range of the above-mentioned low-frequency signal sprayed into the microphone is within the time domain range corresponding to the above-mentioned voice signal, it can be further confirmed that the above-mentioned low-frequency signal sprayed into the microphone is generated by real person voice; if the time domain range corresponding to the above-mentioned low-frequency signal sprayed into the microphone is not within the time domain range of the above-mentioned voice signal, then the above-mentioned low-frequency signal sprayed into the microphone may be generated by wind impacting the microphone in a wind noise scene, and at this time it can be determined that the voiceprint verification of the voice signal has failed.

[0086] In an embodiment of the present application, on the basis that the above-mentioned voice signal includes the above-mentioned low-frequency signal of the microphone, the problem of misidentification of breath detection in wind noise scenarios can be avoided by comparing the low-frequency signal of the microphone with the voice signal in the time domain, thereby further reducing the risk of voiceprint impersonation.

[0087] In the voiceprint verification method provided in the embodiment of the present application, since the voiceprint feature information of the voice signal used for voiceprint verification matches the preset voiceprint feature information, it is possible to first detect whether the voice signal contains a low-frequency signal of a microphone spray, and the low-frequency signal of a microphone spray is usually generated by a real person's voice. Therefore, if the voice signal contains the low-frequency signal of a microphone spray, it is determined that the voiceprint verification of the voice signal is successful. On the premise of ensuring that the voice signal is a voice signal generated by a real person rather than a recorded or reproduced sound, it is determined that the voiceprint verification can be successfully passed, thereby avoiding the misuse of recorded or reproduced sounds, thereby improving the security of voiceprint verification.

[0088] Optionally, in the embodiment of the present application, combined with Figure 1 ,like Fig. 9As shown, before the above step 103, the voiceprint verification method provided in the embodiment of the present application may further include the following step 104; and the step 103 may be specifically implemented by the following step 103b.

[0089] Step 104: The electronic device obtains speech intensity information of the speech signal.

[0090] Optionally, in an embodiment of the present application, the electronic device may calculate the intensity of the breath sound in the pronunciation corresponding to the voice signal through the above-mentioned voiceprint verification model.

[0091] Step 103b: When the voice signal includes a microphone low-frequency signal and the voice intensity information matches the preset voice intensity information, the electronic device determines that the voiceprint verification of the voice signal is successful.

[0092] Optionally, in an embodiment of the present application, the preset voice intensity information may be: during the voiceprint pre-training, the voice intensity information of the user's voice signal during the training process is calculated and saved, and matched one-to-one with the corresponding voice pronunciation.

[0093] Optionally, in an embodiment of the present application, the above-mentioned voice intensity information matches the above-mentioned preset voice intensity information, which can be understood as: the matching degree or similarity between the voice intensity information and the preset voice intensity information is greater than or equal to a second threshold; wherein, the second threshold can be preset by the system, or can be arbitrarily set by the user according to actual usage requirements.

[0094] For example, Fig.10 As shown, the voice pronunciation 10 is "I am the owner of this machine". After acquiring the voice intensity information 11 of the voice signal, the electronic device can match the voice pronunciation 10 with the voice intensity information 11.

[0095] It should be noted that even if some special people can imitate the timbre, due to different pronunciation habits, there will still be differences in breath strength when different people make the same sound. By comparing these differences, it can be used to determine whether these breaths are emitted by the user himself.

[0096] Optionally, in an embodiment of the present application, when the voice intensity information matches the preset voice intensity information, it can be considered that the voice signal is sent by a real user, so that it can be determined that the voiceprint verification of the voice signal is successful.

[0097] Optionally, in an embodiment of the present application, when the voice intensity information does not match the preset voice intensity information, the electronic device may determine that the voiceprint verification of the voice signal has failed.

[0098] Optionally, in the embodiment of the present application, the fact that the voice intensity information does not match the preset voice intensity information may be understood as: a matching degree or similarity between the voice intensity information and the preset voice intensity information is less than the second threshold.

[0099] In an embodiment of the present application, on the basis that the above-mentioned voice signal includes the above-mentioned low-frequency signal of the microphone, by comparing the above-mentioned voice intensity information with the above-mentioned preset voice intensity information, the risk of special persons impersonating voiceprints by imitating timbre can be effectively reduced.

[0100] Optionally, combined Figure 1 ,like Fig.11 As shown, the voiceprint verification method provided in the embodiment of the present application may also include the following step 105.

[0101] Step 105: When the voice signal does not contain the microphone low-frequency signal, the electronic device determines that the voiceprint verification of the voice signal has failed.

[0102] Optionally, in an embodiment of the present application, when the voice signal does not include the microphone low-frequency signal, it can be considered that the voice signal is not a voice signal uttered by a real person, and thus it can be directly determined that the voiceprint verification of the voice signal has failed.

[0103] In the embodiment of the present application, since the electronic device can determine that the voiceprint verification of the voice signal has failed when the voice signal does not contain the low-frequency signal, it can effectively avoid voiceprint impersonation by non-real people, thereby further improving the security of voiceprint verification.

[0104] The voiceprint verification method provided in the embodiment of the present application is exemplarily described below in conjunction with the accompanying drawings.

[0105] For example, Fig.12 As shown, the voiceprint verification method provided in the embodiment of the present application may include the following steps A to C.

[0106] Step A: The user performs voiceprint pre-training.

[0107] Through the voice signal recorded in advance by the user, the voice feature information of the user's voice signal is extracted, and a voiceprint model is generated. The model serves as the basis for subsequent voiceprint verification.

[0108] Step B: The user performs the voiceprint verification process.

[0109] The user needs to place the microphone about 5cm in front of the mouth and speak the corresponding words according to the voiceprint verification requirements, that is, the voice signal used for voiceprint verification, so as to obtain the voice data for this voiceprint verification. The words can be fixed words or random words;

[0110] The voice data of the user's voice signal is extracted and sent to the voiceprint verification algorithm for detection. The voiceprint verification algorithm will extract the voiceprint feature information and compare it with the pre-trained voiceprint model in A to determine whether the voiceprint feature information is consistent with the user's own characteristics. If not, the process ends; if not, the breath detection process begins.

[0111] Step C: Obtain the user's voice signal and perform breath detection process.

[0112] Use the above breath detection algorithm to detect the user's voice signal. If the breath detection fails, it is considered that the user's voice is not a real person, the voiceprint verification process fails, and the process ends; if the breath detection succeeds, it is considered that the voiceprint verification process has passed.

[0113] The above-mentioned method embodiments, or various possible implementation methods in each method embodiment, can be executed separately, or, under the premise that there is no contradiction, can also be executed in combination with each other. The specific implementation can be determined according to actual usage requirements, and the embodiments of the present application do not limit this.

[0114] The voiceprint verification method provided in the embodiment of the present application can be executed by a voiceprint verification device. In the embodiment of the present application, a voiceprint verification device executing the voiceprint verification method is taken as an example to illustrate the voiceprint verification device provided in the embodiment of the present application.

[0115] like Fig.13 As shown, an embodiment of the present application provides a voiceprint verification device 130 , which may include an acquisition module 131 , a detection module 132 and a determination module 133 .

[0116] The acquisition module 131 can be used to collect a voice signal for voiceprint verification and obtain voiceprint feature information of the voice signal. The detection module 132 can be used to perform a detection of a microphone low-frequency signal on the voice signal when the voiceprint feature information matches the preset voiceprint feature information. The determination module 133 can be used to determine that the voiceprint verification of the voice signal is successful when the voice signal contains the microphone low-frequency signal.

[0117] In a possible implementation, the determination module 133 can be specifically used to determine that the voiceprint verification of the voice signal is successful when the voice signal contains the microphone low-frequency signal and the time domain range of the microphone low-frequency signal is within the time domain range of the voice signal.

[0118] In a possible implementation, the acquisition module 131 may also be used to acquire the voice strength information of the voice signal before the determination module 133 determines that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone-spraying low-frequency signal. The determination module 133 may be specifically used to determine that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone-spraying low-frequency signal and the voice strength information matches the preset voice strength information.

[0119] In one possible implementation, the detection module 132 can be specifically used to perform detection of the microphone low-frequency signal on the above-mentioned speech signal through a breath detection model; wherein the breath detection model is a model obtained by training a Gaussian mixture model or a deep learning model using at least two microphone speech signals.

[0120] In a possible implementation, the determination module 133 may also be used to determine that the voiceprint verification of the voice signal has failed when the voice signal does not include the microphone low-frequency signal.

[0121] In the voiceprint verification device provided in the embodiment of the present application, since the voiceprint feature information of the voice signal used for voiceprint verification matches the preset voiceprint feature information, the voiceprint verification device can first detect whether the voice signal contains a low-frequency signal of a microphone, and the low-frequency signal of a microphone is usually generated by a real person's voice. Therefore, if the voice signal contains the low-frequency signal of a microphone, it is determined that the voiceprint verification of the voice signal is successful. On the premise of ensuring that the voice signal is a voice signal generated by a real person rather than a recorded or reproduced sound, it can be determined that the voiceprint verification can be successfully passed, thereby avoiding the misuse of recorded or reproduced sounds, thereby improving the security of voiceprint verification.

[0122] The voiceprint verification device in the embodiment of the present application can be an electronic device, or a component in the electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or other devices other than a terminal. Exemplarily, the electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) device, a robot, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc. It can also be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0123] The voiceprint verification device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an IOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0124] The voiceprint verification device provided in the embodiment of the present application can implement each process implemented in the above method embodiment and achieve the same technical effect. To avoid repetition, it will not be described here.

[0125] like Fig.14 As shown, an embodiment of the present application further provides an electronic device 100, including a processor 101 and a memory 102, wherein the memory 102 stores programs or instructions that can be executed on the processor 101, and when the program or instructions are executed by the processor 101, the various steps of the voiceprint verification method embodiment described above are implemented, and the same technical effect can be achieved. To avoid repetition, they will not be described here.

[0126] It should be noted that the electronic devices in the embodiments of the present application include mobile electronic devices and non-mobile electronic devices.

[0127] Fig.15 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application.

[0128] like Fig.15As shown, the electronic device 1000 includes but is not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and a processor 1010 and other components.

[0129] Those skilled in the art will appreciate that the electronic device 1000 may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the processor 1010 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Fig.15 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0130] The input unit 1004 can be used to collect a voice signal for voiceprint verification and obtain voiceprint feature information of the voice signal. The processor 1010 can be used to perform a detection of a low-frequency signal of a microphone spray on the voice signal when the voiceprint feature information matches the preset voiceprint feature information; and determine that the voiceprint verification of the voice signal is successful when the voice signal contains the low-frequency signal of a microphone spray.

[0131] In one possible implementation, the processor 1010 may be specifically configured to determine that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone low-frequency signal and the time domain range of the microphone low-frequency signal is within the time domain range of the voice signal.

[0132] In a possible implementation, the processor 1010 may also be used to obtain the voice strength information of the voice signal before determining that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone-spraying low-frequency signal. The processor 1010 may be specifically used to determine that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone-spraying low-frequency signal and the voice strength information matches the preset voice strength information.

[0133] In one possible implementation, the processor 1010 can be specifically used to perform detection of the microphone low-frequency signal on the above-mentioned speech signal through a breath detection model; wherein the breath detection model is a model obtained by training a Gaussian mixture model or a deep learning model using at least two microphone speech signals.

[0134] In a possible implementation, the processor 1010 may also be configured to determine that the voiceprint verification of the voice signal has failed when the voice signal does not include the microphone low-frequency signal.

[0135] In the voiceprint verification device provided in the embodiment of the present application, since the voiceprint feature information of the voice signal used for voiceprint verification matches the preset voiceprint feature information, the electronic device can first detect whether the voice signal contains a low-frequency signal of a microphone, and the low-frequency signal of a microphone is usually generated by a real person's voice. Therefore, if the voice signal contains the low-frequency signal of a microphone, it is determined that the voiceprint verification of the voice signal is successful. On the premise of ensuring that the voice signal is a voice signal generated by a real person rather than a recorded or reproduced sound, it can be determined that the voiceprint verification can be successfully passed, thereby avoiding the misuse of recorded or reproduced sounds, thereby improving the security of voiceprint verification.

[0136] It should be understood that in the embodiment of the present application, the input unit 1004 may include a graphics processor (Graphics Processing Unit, GPU) 10041 and a microphone 10042, and the graphics processor 10041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode. The display unit 1006 may include a display panel 10061, and the display panel 10061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1007 includes a touch panel 10071 and at least one of other input devices 10072. The touch panel 10071 is also called a touch screen. The touch panel 10071 may include two parts: a touch detection device and a touch controller. Other input devices 10072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0137] The memory 1009 can be used to store software programs and various data. The memory 1009 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory 1009 may include a volatile memory or a non-volatile memory, or the memory 1009 may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (DRRAM). The memory 1009 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0138] The processor 1010 may include one or more processing units; optionally, the processor 1010 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 1010.

[0139] The embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, each process of the above-mentioned voiceprint verification method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0140] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0141] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned voiceprint verification method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0142] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0143] The embodiment of the present application provides a computer program / program product, which is stored in a storage medium. The program / program product is executed by at least one processor to implement the various processes of the above-mentioned voiceprint verification method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0144] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0145] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0146] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A voiceprint verification method, characterized in that: The method comprises: Collecting a voice signal for voiceprint verification and obtaining voiceprint feature information of the voice signal; In the case where the voiceprint feature information matches the preset voiceprint feature information, performing microphone low-frequency signal detection on the voice signal; In the case that the voice signal includes the microphone low-frequency signal, it is determined that the voiceprint verification of the voice signal is successful.

2. The method according to claim 1, characterized in that When the voice signal includes the microphone low-frequency signal, determining that the voiceprint verification of the voice signal is successful includes: In the case that the voice signal includes the microphone-spraying low-frequency signal, and the time domain range of the microphone-spraying low-frequency signal is within the time domain range of the voice signal, it is determined that the voiceprint verification of the voice signal is successful.

3. The method according to claim 1 or 2, characterized in that: In the case where the voice signal includes the microphone spray low-frequency signal, before determining that the voiceprint verification of the voice signal is successful, the method further includes: Acquiring speech intensity information of the speech signal; When the voice signal includes the microphone spray low-frequency signal, determining that the voiceprint verification of the voice signal is successful includes: When the voice signal includes the microphone low-frequency signal and the voice intensity information matches the preset voice intensity information, it is determined that the voiceprint verification of the voice signal is successful.

4. The method according to claim 1, characterized in that: The detecting of the microphone-spraying low-frequency signal on the voice signal comprises: Performing detection of a low-frequency signal of the microphone on the speech signal through a breath detection model; The breath detection model is a model obtained by training a Gaussian mixture model or a deep learning model using at least two microphone speech signals.

5. The method according to any one of claims 1 to 4, characterized in that The method further comprises: When the voice signal does not include the microphone low-frequency signal, it is determined that the voiceprint verification of the voice signal has failed.

6. A voiceprint verification device, characterized in that: The device comprises an acquisition module, a detection module and a determination module; The acquisition module is used to collect a voice signal for voiceprint verification and obtain voiceprint feature information of the voice signal; The detection module is used to perform microphone low-frequency signal detection on the voice signal when the voiceprint feature information matches the preset voiceprint feature information; The determination module is used to determine that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone low-frequency signal.

7. The device according to claim 6, characterized in that The determination module is specifically used to determine that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone-spraying low-frequency signal and the time domain range of the microphone-spraying low-frequency signal is within the time domain range corresponding to the voice signal.

8. The device according to claim 6 or 7, characterized in that The acquisition module is further configured to acquire the voice strength information of the voice signal before the determination module determines that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone spraying low-frequency signal; The determination module is specifically used to determine that the voiceprint verification of the voice signal is successful when the voice signal includes the microphone low-frequency signal and the voice intensity information matches the preset voice intensity information.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the voiceprint verification method according to any one of claims 1 to 5 are implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the voiceprint verification method according to any one of claims 1 to 5 are implemented.

Images