Voiceprint recognition method and device, computer device and storage medium
By collecting voiceprint signals using a microphone array and utilizing a trained partial discharge signal recognition model, combined with wavelet transform, noise reduction, and phase coding techniques, the accuracy of partial discharge detection of live equipment is improved, solving the problem of inaccurate identification due to noise in traditional methods.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA SOUTHERN POWER GRID GENERAL AVIATION SERVICE CO LTD
- Filing Date
- 2022-11-10
- Publication Date
- 2026-05-19
AI Technical Summary
Traditional voiceprint signal recognition methods are not very accurate in detecting partial discharge in live equipment, mainly because they do not take into account the interference noise in the voiceprint signal.
A microphone array is used to collect voiceprint signals, which are then identified using a trained partial discharge signal recognition model. During model training, the first sample voiceprint signal undergoes a second wavelet transform, and the second sample voiceprint signal undergoes wavelet packet denoising and reconstruction. Combined with signal enhancement and phase coding techniques, the signal quality is improved.
It improves the accuracy of partial discharge signal identification and solves the problem of inaccurate identification caused by not considering interference noise in traditional technology.
Smart Images
Figure CN115762532B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of signal processing technology, and in particular to a voiceprint recognition method, apparatus, computer device, and storage medium. Background Technology
[0002] Partial discharge is a common phenomenon in live electrical equipment, which can lead to damage to the equipment, power outages, and property loss. Therefore, routine inspection for partial discharge in live electrical equipment is a crucial step in fault screening.
[0003] In traditional technologies, routine partial discharge inspections of live equipment are conducted by collecting acoustic signature signals and identifying the presence of partial discharge signals within these signals to determine if partial discharge has occurred. However, the inventors discovered that the accuracy of traditional acoustic signature recognition methods is not high. Summary of the Invention
[0004] Therefore, it is necessary to provide a voiceprint recognition method, apparatus, computer equipment, and storage medium that can improve the accuracy of acquiring partial discharge signals in voiceprint signals, in order to address the above-mentioned technical problems.
[0005] Firstly, this application provides a voiceprint recognition method. The method includes:
[0006] Acquire voiceprint signals collected through a microphone array;
[0007] The voiceprint signal is identified using a trained partial discharge signal recognition model to obtain a voiceprint signal recognition result, which includes a judgment result on whether the voiceprint signal includes a partial discharge signal.
[0008] The methods for training the partial discharge signal recognition model include:
[0009] Acquire the first sample voiceprint signal, and perform a second wavelet transform on the first sample voiceprint signal to obtain the processed first sample voiceprint signal.
[0010] The second sample voiceprint signal is acquired, and wavelet packet denoising and reconstruction processing is performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal.
[0011] The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
[0012] In one embodiment, before training the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal, the method further includes:
[0013] The processed first sample voiceprint signal and the processed second sample voiceprint signal are subjected to signal enhancement processing respectively to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
[0014] In one embodiment, before training the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal, the method further includes:
[0015] Phase encoding is performed on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing, respectively, to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
[0016] In one embodiment, the signal enhancement processing performed on the processed first sample voiceprint signal and the processed second sample voiceprint signal includes:
[0017] The phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are respectively subjected to superposition and averaging processing.
[0018] In one embodiment, performing a second wavelet transform on the first sample voiceprint signal includes:
[0019] The target frequency band signal is extracted from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
[0020] In one embodiment, the wavelet packet denoising and reconstruction processing of the second sample voiceprint signal includes:
[0021] The second sample voiceprint signal is decomposed by wavelet packet decomposition to obtain the decomposed second sample voiceprint signal;
[0022] The decomposed second sample voiceprint signal is filtered according to a preset dynamic threshold to obtain the filtered second sample voiceprint signal.
[0023] The second sample voiceprint signal after screening is reconstructed using the optimal wavelet envelope basis algorithm.
[0024] Secondly, this application also provides a voiceprint recognition device. The device includes:
[0025] The signal acquisition module is used to acquire the voiceprint signal collected by the microphone array;
[0026] The signal recognition module is used to recognize the voiceprint signal using a trained partial discharge signal recognition model, and obtain the voiceprint signal recognition result, the recognition result including a judgment result on whether the voiceprint signal includes a partial discharge signal;
[0027] The methods for training the partial discharge signal recognition model include:
[0028] Acquire the first sample voiceprint signal, and perform a second wavelet transform on the first sample voiceprint signal to obtain the processed first sample voiceprint signal.
[0029] The second sample voiceprint signal is acquired, and wavelet packet denoising and reconstruction processing is performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal.
[0030] The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
[0031] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the methods in any of the above embodiments.
[0032] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the methods in any of the above embodiments.
[0033] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the methods in any of the above embodiments.
[0034] The aforementioned voiceprint recognition method, apparatus, computer equipment, and storage medium acquire voiceprint signals collected through a microphone array; use a trained partial discharge signal recognition model to recognize the voiceprint signals and obtain voiceprint signal recognition results, including a judgment result on whether the voiceprint signal includes a partial discharge signal; the method for training the partial discharge signal recognition model is as follows: acquire a first sample voiceprint signal, perform a second wavelet transform on the first sample voiceprint signal to obtain a processed first sample voiceprint signal; acquire a second sample voiceprint signal, perform wavelet packet denoising and reconstruction processing on the second sample voiceprint signal to obtain a processed second sample voiceprint signal; and train the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal to obtain a trained partial discharge signal recognition model. Compared to traditional techniques that fail to consider interference noise in the voiceprint signal, resulting in inaccurate partial discharge signal identification, this embodiment uses a trained partial discharge signal identification model to identify the voiceprint signal. This model considers both a first sample voiceprint signal and a second sample voiceprint signal, which are different voiceprint signals collected in two different scenarios. Different noise reduction methods are used for these two voiceprint signals, thus solving the problem of low accuracy in obtaining partial discharge signals in voiceprint signals due to the failure to consider interference noise in the voiceprint signal in traditional techniques. Attached Figure Description
[0035] Figure 1 This is a diagram illustrating the application environment of the voiceprint recognition method provided in the embodiments of this application.
[0036] Figure 2 This is a flowchart illustrating the voiceprint recognition method provided in the embodiments of this application;
[0037] Figure 3 This is a flowchart illustrating the method of training a partial discharge signal recognition model in one embodiment;
[0038] Figure 4 This is a schematic diagram of the process for wavelet packet denoising and reconstruction of the second sample voiceprint signal in one embodiment.
[0039] Figure 5 This is a flowchart illustrating the method for training and obtaining a partial discharge signal recognition model in another embodiment;
[0040] Figure 6 This is a flowchart illustrating the method for training and obtaining a partial discharge signal recognition model in yet another embodiment;
[0041] Figure 7 This is a structural block diagram of a voiceprint recognition device provided in an embodiment of this application;
[0042] Figure 8 This is an internal structure diagram of a computer device provided in an embodiment of this application;
[0043] Figure 9 This is an internal structural diagram of another computer device provided in the embodiments of this application. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0045] The voiceprint recognition method provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed in the cloud or on other network servers. Terminal 102 acquires voiceprint signals collected through a microphone array; the acquired voiceprint signals are uploaded to server 104, where a trained partial discharge signal recognition model is used to identify the voiceprint signals and obtain the voiceprint signal recognition result. Terminal 102 can be, but is not limited to, a contact device, a portable device, an online monitoring device, or a drone-mounted device. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.
[0046] The voiceprint recognition method provided in this application embodiment can also be applied to terminal devices. The terminal device acquires voiceprint signals collected by a microphone array; it uses a trained partial discharge signal recognition model to recognize the voiceprint signals and obtains voiceprint signal recognition results, including a judgment result on whether the voiceprint signal includes a partial discharge signal.
[0047] This voiceprint recognition method can also be applied to servers. The server acquires voiceprint signals collected through a microphone array; using a trained partial discharge signal recognition model, it identifies the voiceprint signals and obtains the voiceprint signal recognition result, which includes a determination of whether the voiceprint signal contains partial discharge signals. The server can obtain the voiceprint signal from the terminal device or through other means, such as from a database or other servers.
[0048] Figure 2 This is a flowchart illustrating a voiceprint recognition method provided in one embodiment of this application. Taking the application of this method to a terminal device or server as an example, it includes the following steps:
[0049] S201, acquire the voiceprint signal collected by the microphone array.
[0050] Microphone arrays refer to systems consisting of at least one acoustic sensor (usually a microphone) used to collect voiceprint signals.
[0051] S202, using the trained partial discharge signal recognition model, the voiceprint signal is recognized to obtain the voiceprint signal recognition result, which includes the judgment result of whether the voiceprint signal includes a partial discharge signal.
[0052] The flowchart illustrating the process of training a partial discharge signal recognition model is as follows: Figure 3 As shown, it includes the following:
[0053] S301, acquire the first sample voiceprint signal, perform a second wavelet transform on the first sample voiceprint signal to obtain the processed first sample voiceprint signal.
[0054] The first sample voiceprint signal refers to the signal collected through a microphone array and processed at the edge device without backend computer support. Edge devices can be contact devices, portable devices, online monitoring devices, or drone-mounted devices.
[0055] In this embodiment, a second wavelet transform is performed on the first sample voiceprint signal, including:
[0056] The target frequency band signal is extracted from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
[0057] In this embodiment, when the main energy distributions of the voiceprint signal and the noise signal are inconsistent, the main features in the voiceprint signal can be effectively extracted, noise signal interference can be eliminated, and timeliness can be guaranteed.
[0058] S302, acquire the second sample voiceprint signal, perform wavelet packet denoising and reconstruction on the second sample voiceprint signal to obtain the processed second sample voiceprint signal.
[0059] The second sample voiceprint signal refers to the signal collected through a microphone array and processed by a back-end computer.
[0060] In this embodiment, a schematic diagram of the wavelet packet denoising and reconstruction process for the second sample voiceprint signal is shown below. Figure 4 As shown, S302 above includes the following:
[0061] S3021, perform wavelet packet decomposition on the second sample voiceprint signal to obtain the decomposed second sample voiceprint signal.
[0062] Among them, wavelet packets can decompose the second sample voiceprint signal into multiple levels, and each level retains all the information of the signal, reducing the loss of effective information contained in the signal processing.
[0063] S3022, the decomposed second sample voiceprint signal is filtered according to the preset dynamic threshold to obtain the filtered second sample voiceprint signal.
[0064] In this embodiment, the wavelet packet coefficients of each node in the decomposed second sample voiceprint signal can be filtered by setting a preset dynamic threshold.
[0065] S3023 uses the optimal wavelet envelope basis algorithm to reconstruct the second sample voiceprint signal after screening.
[0066] In this embodiment, the wavelet packet denoising and reconstruction process of the second sample voiceprint signal preserves the characteristics of the voiceprint signal in each frequency band and has great flexibility in voiceprint signal reconstruction, which can effectively improve the signal-to-noise ratio of the voiceprint signal and highlight the main features.
[0067] S303, based on the processed first sample voiceprint signal and the processed second sample voiceprint signal, train the partial discharge signal recognition model to be trained to obtain the trained partial discharge signal recognition model.
[0068] The voiceprint recognition method provided in this embodiment acquires voiceprint signals collected by a microphone array; uses a trained partial discharge signal recognition model to recognize the voiceprint signals and obtain voiceprint signal recognition results, including a judgment result on whether the voiceprint signal includes a partial discharge signal; the method for training the partial discharge signal recognition model is as follows: acquire a first sample voiceprint signal, perform a second wavelet transform on the first sample voiceprint signal to obtain a processed first sample voiceprint signal; acquire a second sample voiceprint signal, perform wavelet packet denoising and reconstruction processing on the second sample voiceprint signal to obtain a processed second sample voiceprint signal; and train the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal to obtain a trained partial discharge signal recognition model. Compared to traditional techniques that fail to consider interference noise in the voiceprint signal, resulting in inaccurate partial discharge signal identification, this embodiment uses a trained partial discharge signal identification model to identify the voiceprint signal. This model considers both a first sample voiceprint signal and a second sample voiceprint signal, which are different voiceprint signals collected in two different scenarios. Different noise reduction methods are used for these two voiceprint signals, thus solving the problem of low accuracy in obtaining partial discharge signals in voiceprint signals due to the failure to consider interference noise in the voiceprint signal in traditional techniques.
[0069] like Figure 5 As shown, Figure 5 This is a flowchart illustrating the method for training a partial discharge signal recognition model in another embodiment, in which the above... Figure 3 Based on the illustrated embodiment, the above-described step S303 includes steps S3031 and S3032 before step S303. This embodiment includes the following:
[0070] S3031, perform signal enhancement processing on the processed first sample voiceprint signal and the processed second sample voiceprint signal respectively to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
[0071] In this embodiment, signal enhancement processing is performed on the processed first sample voiceprint signal and the processed second sample voiceprint signal, including:
[0072] The phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are respectively subjected to superposition and averaging.
[0073] In this embodiment, the phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are superimposed and averaged to suppress the ambient noise signal in the voiceprint signal.
[0074] S3032, the partial discharge signal recognition model to be trained is trained based on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing, so as to obtain the trained partial discharge signal recognition model.
[0075] like Figure 6 As shown, Figure 6 This is a flowchart illustrating the method for training a partial discharge signal recognition model in yet another embodiment. In this embodiment, in the above... Figure 5 Based on the illustrated embodiment, the above-mentioned step S3032 includes steps S30321 and S30322 before step S3032. This embodiment includes the following:
[0076] S30321, perform phase encoding on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing respectively to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
[0077] In this embodiment, the process of phase encoding the voiceprint signal is illustrated with an example:
[0078] Assuming the voiceprint signal is x=[a,b,c,d,e], first solve the phase code corresponding to the voiceprint signal, as shown in the following equation (1);
[0079]
[0080] In this context, PE1, PE2, PE3, PE4, and PE5 are phase codes; w1, w2, w3, w4, and w5 are weight terms (trainable parameters); and b1, b2, b3, b4, and b5 are bias terms (trainable parameters). i is the voiceprint signal index, T is the length of the discrete voiceprint signal in one period, and N is the number of integer periods in which the voiceprint signal is located.
[0081] The phase code is added to x one by one, as shown in equation (2) below;
[0082]
[0083] Where y is the phase-encoded voiceprint signal.
[0084] S30322, the partial discharge signal recognition model to be trained is trained based on the first sample voiceprint signal after phase encoding and the second sample voiceprint signal after phase encoding, so as to obtain the trained partial discharge signal recognition model.
[0085] The partial discharge signal recognition model to be trained can be an LSTM network model.
[0086] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0087] Based on the same inventive concept, this application also provides a voiceprint recognition device for implementing the voiceprint recognition method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more voiceprint recognition device embodiments provided below can be found in the limitations of the voiceprint recognition method described above, and will not be repeated here.
[0088] See Figure 7 , Figure 7 This is a structural block diagram of a voiceprint recognition device provided in an embodiment of this application. The device 700 includes: a signal acquisition module 701, a signal recognition module 702, and a model training module 703, wherein:
[0089] The signal acquisition module 701 is used to acquire the voiceprint signal collected by the microphone array;
[0090] The signal recognition module 702 is used to recognize the voiceprint signal using a trained partial discharge signal recognition model, and obtain the voiceprint signal recognition result, which includes the judgment result of whether the voiceprint signal includes a partial discharge signal.
[0091] Model training module 703 is used to train a partial discharge signal recognition model;
[0092] The model training module 703 includes a first sample processing unit, a second sample processing unit, and a model training unit.
[0093] The first sample processing unit is used to acquire the first sample voiceprint signal, perform a second wavelet transform on the first sample voiceprint signal, and obtain the processed first sample voiceprint signal.
[0094] The second sample processing unit is used to acquire the second sample voiceprint signal, perform wavelet packet denoising and reconstruction processing on the second sample voiceprint signal, and obtain the processed second sample voiceprint signal.
[0095] The model training unit is used to train the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal, so as to obtain the trained partial discharge signal recognition model.
[0096] The voiceprint recognition device provided in this embodiment acquires voiceprint signals collected by a microphone array; it then uses a trained partial discharge signal recognition model to recognize the voiceprint signals and obtain voiceprint signal recognition results, including a judgment result on whether the voiceprint signal includes a partial discharge signal; the method for training the partial discharge signal recognition model is as follows: acquiring a first sample voiceprint signal, performing a second wavelet transform on the first sample voiceprint signal to obtain a processed first sample voiceprint signal; acquiring a second sample voiceprint signal, performing wavelet packet denoising and reconstruction processing on the second sample voiceprint signal to obtain a processed second sample voiceprint signal; and training the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal to obtain a trained partial discharge signal recognition model. Compared to traditional techniques that fail to consider interference noise in the voiceprint signal, resulting in inaccurate partial discharge signal identification, this embodiment uses a trained partial discharge signal identification model to identify the voiceprint signal. This model considers both a first sample voiceprint signal and a second sample voiceprint signal, which are different voiceprint signals collected in two different scenarios. Different noise reduction methods are used for these two voiceprint signals, thus solving the problem of low accuracy in obtaining partial discharge signals in voiceprint signals due to the failure to consider interference noise in the voiceprint signal in traditional techniques.
[0097] Optionally, the device 700 also includes:
[0098] The signal enhancement module is used to perform signal enhancement processing on the processed first sample voiceprint signal and the processed second sample voiceprint signal respectively, so as to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
[0099] Optionally, the device 700 also includes:
[0100] The phase encoding module is used to perform phase encoding on the first sample voiceprint signal and the second sample voiceprint signal after signal enhancement processing, respectively, to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
[0101] Optionally, the signal enhancement module includes:
[0102] The signal enhancement unit is used to perform superposition and averaging processing on the phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal, respectively.
[0103] Optionally, the first sample processing unit includes:
[0104] The first sample processing subunit is used to extract the target frequency band signal from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
[0105] Optionally, the second sample processing unit includes:
[0106] The signal decomposition subunit is used to perform wavelet packet decomposition on the second sample voiceprint signal to obtain the decomposed second sample voiceprint signal.
[0107] The signal filtering subunit is used to filter the decomposed second sample voiceprint signal according to a preset dynamic threshold to obtain the filtered second sample voiceprint signal.
[0108] The signal reconstruction subunit is used to reconstruct the filtered second sample voiceprint signal using the optimal wavelet envelope basis algorithm.
[0109] Each module in the aforementioned voiceprint recognition device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.
[0110] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores voiceprint signals and recognition result data. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a voiceprint recognition method.
[0111] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a voiceprint recognition method. The display screen can be an LCD screen or an e-ink display screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.
[0112] Those skilled in the art will understand that Figure 8 and Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0113] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the voiceprint recognition method provided in the above embodiment.
[0114] Acquire voiceprint signals collected through a microphone array;
[0115] Using a trained partial discharge signal recognition model, the voiceprint signal is recognized to obtain the voiceprint signal recognition result, which includes the judgment result of whether the voiceprint signal includes a partial discharge signal.
[0116] Methods for training a partial discharge signal recognition model include:
[0117] The first sample voiceprint signal is acquired, and a second wavelet transform is performed on the first sample voiceprint signal to obtain the processed first sample voiceprint signal.
[0118] The second sample voiceprint signal is obtained, and wavelet packet denoising and reconstruction are performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal.
[0119] The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
[0120] In one embodiment, the processor, when executing a computer program, also performs the following steps:
[0121] The processed first sample voiceprint signal and the processed second sample voiceprint signal are subjected to signal enhancement processing respectively to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
[0122] In one embodiment, the processor, when executing a computer program, also performs the following steps:
[0123] Phase encoding is performed on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing, respectively, to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
[0124] In one embodiment, the processor, when executing a computer program, also performs the following steps:
[0125] The phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are respectively subjected to superposition and averaging.
[0126] In one embodiment, the processor, when executing a computer program, also performs the following steps:
[0127] The target frequency band signal is extracted from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
[0128] In one embodiment, the processor, when executing a computer program, also performs the following steps:
[0129] Wavelet packet decomposition is performed on the second sample voiceprint signal to obtain the decomposed second sample voiceprint signal;
[0130] The decomposed second sample voiceprint signal is filtered according to a preset dynamic threshold to obtain the filtered second sample voiceprint signal.
[0131] The optimal wavelet envelope basis algorithm is used to reconstruct the second sample voiceprint signal after screening.
[0132] The implementation principle and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.
[0133] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, it implements the steps of the voiceprint signal recognition method provided in the above embodiment:
[0134] Acquire voiceprint signals collected through a microphone array;
[0135] Using a trained partial discharge signal recognition model, the voiceprint signal is recognized to obtain the voiceprint signal recognition result, which includes the judgment result of whether the voiceprint signal includes a partial discharge signal.
[0136] Methods for training a partial discharge signal recognition model include:
[0137] The first sample voiceprint signal is acquired, and a second wavelet transform is performed on the first sample voiceprint signal to obtain the processed first sample voiceprint signal.
[0138] The second sample voiceprint signal is obtained, and wavelet packet denoising and reconstruction are performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal.
[0139] The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
[0140] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0141] The processed first sample voiceprint signal and the processed second sample voiceprint signal are subjected to signal enhancement processing respectively to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
[0142] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0143] Phase encoding is performed on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing, respectively, to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
[0144] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0145] The phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are respectively subjected to superposition and averaging.
[0146] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0147] The target frequency band signal is extracted from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
[0148] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0149] Wavelet packet decomposition is performed on the second sample voiceprint signal to obtain the decomposed second sample voiceprint signal;
[0150] The decomposed second sample voiceprint signal is filtered according to a preset dynamic threshold to obtain the filtered second sample voiceprint signal.
[0151] The optimal wavelet envelope basis algorithm is used to reconstruct the second sample voiceprint signal after screening.
[0152] The implementation principle and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.
[0153] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the voiceprint signal recognition method provided in the above embodiment:
[0154] Acquire voiceprint signals collected through a microphone array;
[0155] Using a trained partial discharge signal recognition model, the voiceprint signal is recognized to obtain the voiceprint signal recognition result, which includes the judgment result of whether the voiceprint signal includes a partial discharge signal.
[0156] Methods for training a partial discharge signal recognition model include:
[0157] The first sample voiceprint signal is acquired, and a second wavelet transform is performed on the first sample voiceprint signal to obtain the processed first sample voiceprint signal.
[0158] The second sample voiceprint signal is obtained, and wavelet packet denoising and reconstruction are performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal.
[0159] The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
[0160] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0161] The processed first sample voiceprint signal and the processed second sample voiceprint signal are subjected to signal enhancement processing respectively to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
[0162] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0163] Phase encoding is performed on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing, respectively, to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
[0164] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0165] The phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are respectively subjected to superposition and averaging.
[0166] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0167] The target frequency band signal is extracted from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
[0168] In one embodiment, when the computer program is executed by a processor, it further performs the following steps:
[0169] Wavelet packet decomposition is performed on the second sample voiceprint signal to obtain the decomposed second sample voiceprint signal;
[0170] The decomposed second sample voiceprint signal is filtered according to a preset dynamic threshold to obtain the filtered second sample voiceprint signal.
[0171] The optimal wavelet envelope basis algorithm is used to reconstruct the second sample voiceprint signal after screening.
[0172] The implementation principle and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.
[0173] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0174] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0175] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A voiceprint recognition method, characterized in that, The method includes: Acquire voiceprint signals collected through a microphone array; The voiceprint signal is identified using a trained partial discharge signal recognition model to obtain a voiceprint signal recognition result, which includes a judgment result on whether the voiceprint signal includes a partial discharge signal. The methods for training the partial discharge signal recognition model include: Acquire the first sample voiceprint signal, and perform a second wavelet transform on the first sample voiceprint signal to obtain the processed first sample voiceprint signal. The second sample voiceprint signal is acquired, and wavelet packet denoising and reconstruction processing is performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal. The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
2. The method according to claim 1, characterized in that, Before training the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal, the method further includes: The processed first sample voiceprint signal and the processed second sample voiceprint signal are subjected to signal enhancement processing respectively to obtain the signal-enhanced first sample voiceprint signal and the signal-enhanced second sample voiceprint signal.
3. The method according to claim 2, characterized in that, Before training the partial discharge signal recognition model to be trained based on the processed first sample voiceprint signal and the processed second sample voiceprint signal, the method further includes: Phase encoding is performed on the first sample voiceprint signal after signal enhancement processing and the second sample voiceprint signal after signal enhancement processing, respectively, to obtain the phase-encoded first sample voiceprint signal and the phase-encoded second sample voiceprint signal.
4. The method according to claim 2 or 3, characterized in that, The step of performing signal enhancement processing on the processed first sample voiceprint signal and the processed second sample voiceprint signal includes: The phase shift signals in the processed first sample voiceprint signal and the processed second sample voiceprint signal are respectively subjected to superposition and averaging processing.
5. The method according to any one of claims 1-3, characterized in that, The step of performing a second wavelet transform on the first sample voiceprint signal includes: The target frequency band signal is extracted from the first sample voiceprint signal by performing a second wavelet transform on the first sample voiceprint signal.
6. The method according to any one of claims 1-3, characterized in that, The wavelet packet denoising and reconstruction processing of the second sample voiceprint signal includes: The second sample voiceprint signal is decomposed by wavelet packet decomposition to obtain the decomposed second sample voiceprint signal; The decomposed second sample voiceprint signal is filtered according to a preset dynamic threshold to obtain the filtered second sample voiceprint signal. The second sample voiceprint signal after screening is reconstructed using the optimal wavelet envelope basis algorithm.
7. A voiceprint recognition device, characterized in that, The device includes: The signal acquisition module is used to acquire the voiceprint signal collected by the microphone array; The signal recognition module is used to recognize the voiceprint signal using a trained partial discharge signal recognition model, and obtain the voiceprint signal recognition result, the recognition result including a judgment result on whether the voiceprint signal includes a partial discharge signal; The methods for training the partial discharge signal recognition model include: Acquire the first sample voiceprint signal, and perform a second wavelet transform on the first sample voiceprint signal to obtain the processed first sample voiceprint signal. The second sample voiceprint signal is acquired, and wavelet packet denoising and reconstruction processing is performed on the second sample voiceprint signal to obtain the processed second sample voiceprint signal. The partial discharge signal recognition model to be trained is obtained by training the first sample voiceprint signal and the second sample voiceprint signal after processing.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.