Signal howling suppression method, device, computer equipment and storage medium

Abstract

The present application relates to a signal howling suppression method, device, computer equipment and storage medium. The above-mentioned signal howling suppression method, device, computer equipment and storage medium, by processing the acoustic signal and the echo signal to obtain a periodic signal, is conducive to the subsequent signal division, the sub-signals obtained after the division are matched, and the sub-signals are used to match the frames, and the accuracy and robustness of the sub-signal information recognition and matching can be guaranteed even in the case of large environmental noise. Finally, howling detection and howling suppression are performed through sub-signal matching. The complexity of this method is low, which is conducive to the application of real-time systems.

Description

Technical Field

[0001] The present application relates to the technical field of speech signal processing, and in particular to a signal howling suppression method, device, computer equipment and storage medium. Background Art

[0002] As an important medium for information transmission in daily life, sound makes it easy to exchange information between each other. However, the propagation of sound is greatly affected by distance. In order to ensure the normal transmission of information, the sound reinforcement system has entered everyone's field of vision. The sound reinforcement system usually refers to a system that amplifies sound signals in real time. It is mainly composed of speakers, microphones, power amplifiers and other equipment. It is widely used in multimedia teaching, concerts and large-scale evening parties.

[0003] When the sound reinforcement system is used in practice, acoustic feedback will occur, which has a great impact on the performance of the sound reinforcement system. Acoustic feedback refers to the audio signal being output by the speaker and then repeatedly collected by the microphone, forming a positive feedback loop between the speaker-microphone-amplifier-speaker, which continuously amplifies the output signal amplitude of the amplifier, causing the signal at certain frequencies to self-oscillate and produce howling.

[0004] At present, from the perspective of suppressing the occurrence of positive feedback, there are methods to suppress the phase delay conditions of positive feedback, including phase shift method, frequency shift method and random phase method. Such algorithms can achieve good results, but the algorithm complexity is relatively high, which is not conducive to the application of real-time systems. Summary of the invention

[0005] Based on this, it is necessary to provide a signal howling suppression method, device, computer equipment and storage medium to address the above technical problems.

[0006] A signal howling suppression method, the method comprising:

[0007] Acquire an acoustic signal frame sequence and an echo signal frame sequence, and process the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal;

[0008] Dividing the acoustic periodic signal and the echo periodic signal respectively, and matching sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair;

[0009] Performing howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and performing howling suppression on the howling sub-signal matching pair to obtain a howling suppression sub-signal;

[0010] Howling suppression is performed on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

[0011] Specifically, the acoustic periodic signal and the echo periodic signal are divided respectively, and sub-signals in the acoustic periodic signal and the echo periodic signal are matched to obtain a sub-signal matching pair, including:

[0012] Respectively dividing the acoustic periodic signal and the echo periodic signal to obtain a sub-acoustic signal and a sub-echo signal;

[0013] Calculate the first phase change eigenvalue of each sub-sound signal:

[0014]

[0015] Among them, f1 i,j is the first phase change characteristic value of the jth sub-acoustic signal of the i-th frame acoustic signal in the acoustic signal frame sequence, is the phase of the jth sub-sound signal of the i-th frame sound signal, μ and η are both preset values;

[0016] Calculate the second phase change characteristic value of each sub-echo signal:

[0017]

[0018] Among them, f2 i,j is the second phase change characteristic value of the jth sub-echo signal of the i-th frame echo signal in the echo signal frame sequence, is the phase of the j-th sub-echo signal of the i-th frame echo signal;

[0019] Calculate the sum of the differences between the first phase change eigenvalue and the second phase change eigenvalue:

[0020]

[0021] Among them, m is the offset parameter, f1 i1,j represents the first phase change characteristic value of the jth sub-sound signal of the i-th frame sound signal, f2 i1+m,j Represents the second phase change characteristic value of the j-th sub-echo signal of the i1+m-th frame echo signal.

[0022] The minimum difference and the corresponding offset parameter are selected to match the sub-sound signal and the sub-echo signal to obtain a sub-signal matching pair.

[0023] Specifically, performing howling detection on the sub-signal matching pair to obtain the howling sub-signal matching pair includes:

[0024] Calculate the three eigenvalues ​​of the amplitude change between matching pairs of adjacent frame sub-signals:

[0025]

[0026] Among them, {h1 i,j ,h2 i,j ,h3 i,j} are the three eigenvalues ​​of amplitude change, ψ1 i,j is the amplitude variation characteristic value of the sub-sound signal, ψ2 i+m1,j is the amplitude variation characteristic value of the sub-echo signal in the corresponding sub-band waveform matching pair, σ 2 is the proportional parameter, p and q are preset values;

[0027] Howling detection is performed according to the amplitude change characteristic values, and when the three amplitude change characteristic values ​​are all greater than a preset threshold, the corresponding sub-signal matching pairs are marked to obtain the howling sub-signal matching pairs.

[0028] In one embodiment, performing howling suppression on the howling sub-signal matching pair to obtain the howling suppression sub-signal includes:

[0029] Howling suppression is performed on the howling sub-signal matching pair, and the howling suppression sub-signal is obtained as follows:

[0030] newt i,j (k) = α × (t1 i,j (k)-β×t2 i+m1,j (k))

[0031] Among them, newt i,j (k) is the howling suppression sub-signal, k is the frequency of the periodic signal; t1 i,j (k) is the jth sub-sound signal of the i-th frame sound signal, t2 i+m1,j (k) is the jth sub-echo signal of the i1+mth frame echo signal in the corresponding howling sub-signal matching pair, α is the suppression parameter, and β is the fusion parameter.

[0032] In one embodiment, the method further comprises:

[0033] Acquire an acoustic signal and an echo signal, perform AD conversion on the acoustic signal and the echo signal respectively, and then perform frame processing to obtain an acoustic signal frame sequence and an echo signal frame sequence.

[0034] In one embodiment, the acoustic signal frame sequence and the echo signal frame sequence are processed to obtain an acoustic periodic signal and an echo periodic signal, including:

[0035] Perform Fourier transform on the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal.

[0036] In one embodiment, performing howling suppression on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal includes:

[0037] The howling suppression sub-signal is combined with an unprocessed normal sub-signal to obtain a new acoustic periodic signal and an echo periodic signal respectively, and the new acoustic periodic signal and the echo periodic signal are subjected to inverse Fourier transform to obtain a target signal.

[0038] A signal howling suppression device, the device comprising:

[0039] A sequence processing module, used for acquiring an acoustic signal frame sequence and an echo signal frame sequence, and processing the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal;

[0040] A sub-signal matching module, used for dividing the acoustic periodic signal and the echo periodic signal respectively, and matching the sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair;

[0041] a howling suppression module, configured to perform howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and perform howling suppression on the howling sub-signal matching pair to obtain a howling suppression sub-signal;

[0042] The output module is used to perform howling suppression on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

[0043] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0044] Acquire an acoustic signal frame sequence and an echo signal frame sequence, and process the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal;

[0045] Dividing the acoustic periodic signal and the echo periodic signal respectively, and matching sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair;

[0046] Performing howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and performing howling suppression on the howling sub-signal matching pair to obtain a howling suppression sub-signal;

[0047] Howling suppression is performed on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

[0048] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0049] Acquire an acoustic signal frame sequence and an echo signal frame sequence, and process the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal;

[0050] Dividing the acoustic periodic signal and the echo periodic signal respectively, and matching sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair;

[0051] Performing howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and performing howling suppression on the howling sub-signal matching pair to obtain a howling suppression sub-signal;

[0052] Howling suppression is performed on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

[0053] The above-mentioned signal howling suppression method, device, computer equipment and storage medium, by processing the acoustic signal and the echo signal to obtain a periodic signal, is conducive to the subsequent signal division, the sub-signals obtained after the division are matched, and the sub-signals are used to match the frames, and the accuracy and robustness of the sub-signal information recognition and matching can be guaranteed even in the case of large environmental noise, and finally the howling detection and howling suppression are performed through the sub-signal matching. The complexity of this method is low, which is conducive to the application of real-time systems, and can also ensure a high howling detection rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A schematic diagram of a signal howling suppression method in an embodiment;

[0055] Figure 2 Schematic diagram of an application of a signal howling suppression method in an embodiment;

[0056] Figure 3 is a structural block diagram of a signal howling device in an embodiment;

[0057] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] In one embodiment, Figure 1 As shown, a signal howling suppression method is provided, comprising the following steps:

[0060] Step 102: Acquire an acoustic signal frame sequence and an echo signal frame sequence, process the acoustic signal frame sequence and the echo signal frame sequence, and obtain an acoustic periodic signal and an echo periodic signal.

[0061] The acoustic signal refers to the voice signal that the user directly transmits to the microphone (such as a microphone), and the echo signal refers to the user's voice signal transmitted to the loudspeaker via wireless communication (such as Bluetooth), and the loudspeaker amplifies the voice signal and then transmits it back to the microphone via wireless communication.

[0062] Step 104 , dividing the acoustic periodic signal and the echo periodic signal respectively, and matching the sub-signals in the acoustic periodic signal and the echo periodic signal to obtain sub-signal matching pairs.

[0063] The acoustic periodic signal and the echo periodic signal are divided respectively. Assuming that the periodic signal is divided into N sub-signals, the bandwidth of each sub-signal is 1 / N of the bandwidth of the original signal.

[0064] The sub-signals obtained after division are matched, and inter-frame matching is performed through the sub-signals, which can ensure the accuracy and robustness of sub-signal information recognition and matching even in the case of large environmental noise.

[0065] Step 106: perform howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and perform howling suppression on the howling sub-signal matching pair to obtain a howling suppressed sub-signal.

[0066] If the howling signal is not suppressed, its energy will continue to increase, so the amplitude change of the howling sub-signal is much greater than that of other sub-signals. Therefore, we can choose to perform howling detection by calculating the amplitude change characteristics between adjacent M frame signals, and then perform howling detection after inter-frame matching between the two signals. Since the signal information used in the detection is richer, the detection rate is higher and the detection effect is more accurate and reliable, thereby improving the howling suppression effect.

[0067] Step 108: perform howling suppression on the acoustic periodic signal and the echo periodic signal according to the howling suppression sub-signal.

[0068] In the above signal howling suppression method, the acoustic signal and the echo signal are processed to obtain a periodic signal, which is conducive to the subsequent signal division, and the sub-signals obtained after the division are matched, and the sub-signals are used to match the frames, so that the accuracy and robustness of the sub-signal information recognition and matching can be guaranteed even in the case of large environmental noise, and finally the howling detection and howling suppression are performed through the sub-signal matching. The complexity of this method is low, which is conducive to the application of real-time systems, and can also ensure a high howling detection rate.

[0069] In one embodiment, processing the acoustic signal frame sequence and the echo signal frame sequence to obtain the acoustic periodic signal and the echo periodic signal includes: performing Fourier transform on the acoustic signal frame sequence and the echo signal frame sequence to obtain the acoustic periodic signal and the echo periodic signal.

[0070] In one embodiment, the acoustic periodic signal and the echo periodic signal are divided respectively, and sub-signals in the acoustic periodic signal and the echo periodic signal are matched to obtain a sub-signal matching pair, including:

[0071] The acoustic periodic signal and the echo periodic signal are divided respectively to obtain sub-acoustic signals and sub-echo signals;

[0072] Calculate the first phase change eigenvalue of each sub-sound signal:

[0073]

[0074] Among them, f1 i,j is the first phase change characteristic value of the jth sub-acoustic signal of the i-th frame acoustic signal in the acoustic signal frame sequence, is the phase of the jth sub-sound signal of the i-th frame sound signal, μ and η are both preset values;

[0075] Calculate the second phase change characteristic value of each sub-echo signal:

[0076]

[0077] Among them, f2 i,j is the second phase change characteristic value of the jth sub-echo signal of the i-th frame echo signal in the echo signal frame sequence, is the phase of the j-th sub-echo signal of the i-th frame echo signal;

[0078] Calculate the sum of the differences between the first phase change eigenvalue and the second phase change eigenvalue:

[0079]

[0080] Among them, m is the offset parameter, f1 i1,j represents the first phase change characteristic value of the jth sub-sound signal of the i-th frame sound signal, f2 i1+m,j Represents the second phase change characteristic value of the j-th sub-echo signal of the i1+m-th frame echo signal.

[0081] The minimum difference and the corresponding offset parameter are selected to match the sub-sound signal and the sub-echo signal to obtain a sub-signal matching pair.

[0082] In one embodiment, performing howling detection on the sub-signal matching pair to obtain the howling sub-signal matching pair includes:

[0083] Calculate the three eigenvalues ​​of the amplitude change between matching pairs of adjacent frame sub-signals:

[0084] Among them, are the three eigenvalues ​​of amplitude change, ψ1 i,j is the amplitude variation characteristic value of the sub-sound signal, ψ2 i+m1,j is the amplitude variation characteristic value of the sub-echo signal in the corresponding sub-band waveform matching pair, σ 2 is the proportional parameter, p and q are preset values;

[0085] Howling detection is performed according to the amplitude change characteristic values. When the three characteristic values ​​of the amplitude change are all greater than a preset threshold, the corresponding sub-signal matching pairs are marked to obtain the howling sub-signal matching pairs.

[0086] In one embodiment, performing howling suppression on a howling sub-signal matching pair to obtain a howling suppression sub-signal includes:

[0087] Howling suppression is performed on the matching pair of the howling sub-signals, and the howling suppression sub-signal is obtained as follows:

[0088] newt i,j (k) = α × (t1 i,j (k)-β×t2 i+m1,j (k))

[0089] Among them, newt i,j (k) is the howling suppression sub-signal, k is the frequency of the periodic signal; t1 i,j (k) is the jth sub-sound signal of the i-th frame sound signal, t2 i+m1,j (k) is the jth sub-echo signal of the i1+mth frame echo signal in the corresponding howling sub-signal matching pair, α is the suppression parameter, and β is the fusion parameter.

[0090] In one embodiment, the method further includes: acquiring an acoustic signal and an echo signal, performing AD conversion on the acoustic signal and the echo signal respectively and then performing frame processing to obtain an acoustic signal frame sequence and an echo signal frame sequence.

[0091] In one embodiment, performing howling suppression on the acoustic periodic signal and the echo periodic signal according to the howling suppression sub-signal includes:

[0092] The howling suppression sub-signal is combined with the unprocessed normal sub-signal to obtain a new acoustic periodic signal and an echo periodic signal respectively, and the new acoustic periodic signal and the echo periodic signal are subjected to inverse Fourier transform to obtain a target signal.

[0093] Specifically, an embodiment is provided to illustrate the method in detail:

[0094] like Figure 2As shown, the user's voice is first collected through a clip-on wireless microphone, and then transmitted to the loudspeaker via Bluetooth wireless. The loudspeaker amplifies the voice and plays it out, while transmitting the signal back to the microphone via Bluetooth. The processing chip processes the two signals to solve the howling suppression problem and improve the voice playback quality.

[0095] 1. The first signal y1(n) is input by the collar-clip wireless microphone for AD conversion, and the converted digital signal is divided into frames y1 i (n), where i is the frame number and n is the number of sampling points.

[0096] 2. The second signal y2(n) of the loudspeaker is transmitted back to the microphone via Bluetooth. Here, the second signal y2(n) is s milliseconds earlier than the first signal y1(n). s∈(1,20) is a time parameter determined in advance. y2(n) is also framed to obtain y2 i (n), where i is the frame number and n is the number of sampling points.

[0097] 3. For the speech signal y1 i (n) and y2 i (n) respectively perform Fourier transform to obtain Y1 i (k) and Y2 i (k), where k represents the frequency;

[0098] 4. Signal Y1 i (k) and Y2 i (k) is divided into 64 sub-bands, denoted as {t1 i,j (k)|j=1,…,64} and {t2 i,j (k)|j=1,…,64}, the bandwidth of each subband signal is 1 / 64 of the original signal bandwidth;

[0099] 5. Align the subbands of the two speech signals using phase information:

[0100] (5.1) For each subband t1 of signal 1 i,j (k), statistics of its phase information Calculate a phase change feature f1 i,j , the specific formula is as follows:

[0101]

[0102] (5.2) For each subband t2 of signal 2 i,j (k), statistics of its phase information Calculate a phase change feature f2 i,j , the specific formula is as follows:

[0103]

[0104] (5.3) Let the current frame i = i1, and calculate the difference and sum of the phase change characteristics of the corresponding sub-bands of the two signals. The formula is as follows:

[0105]

[0106] Where the parameter m∈[0,100].

[0107] (5.4) Sort p(m) from small to large, and select the parameter m1 corresponding to the minimum value as the registration result of the two signals.

[0108] 6. For each sub-band of the two signals, the amplitude value characteristics are used to perform howling detection;

[0109] (6.1) For each subband t1 of signal 1 i,j (k), count its amplitude information ψ1 i,j ;

[0110] (6.2) For each subband t2 of signal 2 i,j (k), count its amplitude information ψ2 i,j ;

[0111] (6.3) Since the energy of the howling signal will continue to increase if it is not suppressed, the amplitude change of the howling subband is much greater than that of other subbands. The three characteristics of the amplitude change between adjacent frames of signals {h1 i,j ,h2 i,j ,h3 i,j}, the specific formula is as follows:

[0112]

[0113] where σ 2 is the scale parameter;

[0114] (6.4) When condition h1 is satisfied i,j >λ,h2 i,j >λ,h3 i,j >λ, where λ is a preset threshold, and the sub-band is labeled 1, otherwise it is 0; the sub-band labeled 1 is the final howling detection result.

[0115] 7. Multiply the sub-band detected by howling by a smaller coefficient to suppress howling. The formula is as follows:

[0116] newt i,j (k) = α × (t1 i,j (k)-β×t2 i+m1,j (k)) (5)

[0117] Among them, α∈(0,1) is the suppression parameter and β∈(0,1) is the fusion parameter.

[0118] 8. Combine the sub-band after howling suppression and other unprocessed sub-bands (i.e., the sub-band labeled 0) into a new signal newY i (k), through the inverse Fourier transform, we can get the speech signal newy after howling suppression i (n).

[0119] The method has a high howling detection rate, is simple and practical in howling suppression, has a fast calculation speed, can meet the needs of real-time processing, is very suitable for the application market of lavalier wireless microphones, and has high practical value.

[0120] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0121] In one embodiment, Figure 3 As shown, a signal howling suppression device is provided, comprising: a sequence processing module, a sub-signal matching module, a howling suppression module and an output module, wherein:

[0122] A sequence processing module, used for acquiring an acoustic signal frame sequence and an echo signal frame sequence, and processing the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal;

[0123] A sub-signal matching module is used to divide the acoustic periodic signal and the echo periodic signal respectively, and match the sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair;

[0124] A howling suppression module is used to perform howling detection on the sub-signal matching pairs to obtain a howling sub-signal matching pair, and perform howling suppression on the howling sub-signal matching pairs to obtain a howling suppression sub-signal;

[0125] The output module is used to perform howling suppression on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

[0126] In one embodiment, the sequence processing module is further used to perform Fourier transform on the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal.

[0127] In one embodiment, the sub-signal matching module is further used to divide the acoustic periodic signal and the echo periodic signal respectively to obtain a sub-acoustic signal and a sub-echo signal;

[0128] Calculate the first phase change eigenvalue of each sub-sound signal:

[0129]

[0130] Among them, f1 i,j is the first phase change characteristic value of the jth sub-acoustic signal of the i-th frame acoustic signal in the acoustic signal frame sequence, is the phase of the jth sub-sound signal of the i-th frame sound signal, μ and η are both preset values;

[0131] Calculate the second phase change characteristic value of each sub-echo signal:

[0132]

[0133] Among them, f2 i,j is the second phase change characteristic value of the jth sub-echo signal of the i-th frame echo signal in the echo signal frame sequence, is the phase of the j-th sub-echo signal of the i-th frame echo signal;

[0134] Calculate the sum of the differences between the first phase change eigenvalue and the second phase change eigenvalue:

[0135]

[0136] Among them, m is the offset parameter, f1 i1,j represents the first phase change characteristic value of the jth sub-sound signal of the i-th frame sound signal, f2 i1+m,j Represents the second phase change characteristic value of the j-th sub-echo signal of the i1+m-th frame echo signal.

[0137] The minimum difference and the corresponding offset parameter are selected to match the sub-sound signal and the sub-echo signal to obtain a sub-signal matching pair.

[0138] In one embodiment, the howling suppression module is further used to calculate three characteristic values ​​of amplitude changes between matching pairs of adjacent frame sub-signals:

[0139]

[0140] Among them, {h1 i,j ,h2 i,j ,h3 i,j} are the three eigenvalues ​​of amplitude change, ψ1 i,j is the amplitude variation characteristic value of the sub-sound signal, ψ2 i+m1,j is the amplitude variation characteristic value of the sub-echo signal in the corresponding sub-band waveform matching pair, σ 2 is the proportional parameter, p and q are preset values;

[0141] Howling detection is performed according to the amplitude change characteristic values, and when the three characteristic values ​​of the amplitude change are all greater than the preset threshold value, the corresponding sub-signal matching pairs are marked to obtain the howling sub-signal matching pairs;

[0142] Howling suppression is performed on the matching pair of the howling sub-signals, and the howling suppression sub-signal is obtained as follows:

[0143] newt i,j (k) = α × (t1 i,j (k)-β×t2 i+m1,j (k))

[0144] Among them, newt i,j (k) is the howling suppression sub-signal, k is the frequency of the periodic signal; t1 i,j (k) is the jth sub-sound signal of the i-th frame sound signal, t2 i+m1,j (k) is the jth sub-echo signal of the i1+mth frame echo signal in the corresponding howling sub-signal matching pair, α is the suppression parameter, and β is the fusion parameter.

[0145] In one embodiment, the output module is further used to combine the howling suppression sub-signal with the unprocessed normal sub-signal to obtain a new acoustic periodic signal and an echo periodic signal respectively, and perform inverse Fourier transform on the new acoustic periodic signal and the echo periodic signal to obtain a target signal.

[0146] For the specific definition of the signal howling suppression device, please refer to the definition of the signal howling suppression method above, which will not be repeated here. Each module in the above-mentioned signal howling suppression device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0147] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store voice signal data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a signal howling suppression method is implemented.

[0148] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0149] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0150] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0151] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0152] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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.

[0153] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A signal howling suppression method, characterized in that: The method comprises: Acquire an acoustic signal frame sequence and an echo signal frame sequence, and process the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal; Dividing the acoustic periodic signal and the echo periodic signal respectively, and matching sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair; Performing howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and performing howling suppression on the howling sub-signal matching pair to obtain a howling suppression sub-signal; Howling suppression is performed on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

2. The method according to claim 1, characterized in that The acoustic periodic signal and the echo periodic signal are divided respectively, and sub-signals in the acoustic periodic signal and the echo periodic signal are matched to obtain a sub-signal matching pair, including: Respectively dividing the acoustic periodic signal and the echo periodic signal to obtain a sub-acoustic signal and a sub-echo signal; Calculate the first phase change eigenvalue of each sub-sound signal: Among them, f1 i,j is the first phase change characteristic value of the jth sub-acoustic signal of the i-th frame acoustic signal in the acoustic signal frame sequence, is the phase of the jth sub-sound signal of the i-th frame sound signal, μ and η are both preset values; Calculate the second phase change characteristic value of each sub-echo signal: Among them, f2 i,j is the second phase change characteristic value of the jth sub-echo signal of the i-th frame echo signal in the echo signal frame sequence, is the phase of the j-th sub-echo signal of the i-th frame echo signal; Calculate the sum of the differences between the first phase change eigenvalue and the second phase change eigenvalue: Among them, m is the offset parameter, f1 i1,j represents the first phase change characteristic value of the jth sub-sound signal of the i-th frame sound signal, f2 i1+m,j represents the second phase change characteristic value of the j-th sub-echo signal of the i1+m-th frame echo signal; The minimum difference and the corresponding offset parameter are selected to match the sub-sound signal and the sub-echo signal to obtain a sub-signal matching pair.

3. The method according to claim 2, characterized in that The performing howling detection on the sub-signal matching pair to obtain the howling sub-signal matching pair includes: Calculate the three eigenvalues ​​of the amplitude change between matching pairs of adjacent frame sub-signals: Among them, {h1 i,j ,h2 i,j ,h3 i,j } are the three eigenvalues ​​of amplitude change, ψ1i,j is the amplitude change eigenvalue of the sub-sound signal, ψ2 i+m1,j is the amplitude variation characteristic value of the sub-echo signal in the corresponding sub-band waveform matching pair, σ 2 is the proportional parameter, p and q are preset values; Howling detection is performed according to the amplitude change characteristic values, and when the three characteristic values ​​of the amplitude change are all greater than a preset threshold, the corresponding sub-signal matching pairs are marked to obtain the howling sub-signal matching pairs.

4. The method according to claim 1, characterized in that: The performing howling suppression on the howling sub-signal matching pair to obtain the howling suppression sub-signal includes: Howling suppression is performed on the howling sub-signal matching pair, and the howling suppression sub-signal is obtained as follows: newt i,j (k)=α×(t1 i,j (k)-β×t2 i+m1,j (k)) Among them, newt i,j (k) is the howling suppression sub-signal, k is the frequency of the periodic signal; t1 i,j (k) is the jth sub-sound signal of the i-th frame sound signal, t2 i+m1,j (k) is the jth sub-echo signal of the i1+mth frame echo signal in the corresponding howling sub-signal matching pair, α is the suppression parameter, and β is the fusion parameter.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Acquire an acoustic signal and an echo signal, perform AD conversion on the acoustic signal and the echo signal respectively, and then perform frame processing to obtain an acoustic signal frame sequence and an echo signal frame sequence.

6. The method according to claim 1, characterized in that Processing the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal, including: Perform Fourier transform on the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal.

7. The method according to claim 1, characterized in that The step of performing howling suppression on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal includes: The howling suppression sub-signal is combined with an unprocessed normal sub-signal to obtain a new acoustic periodic signal and an echo periodic signal respectively, and the new acoustic periodic signal and the echo periodic signal are subjected to inverse Fourier transform to obtain a target signal.

8. A signal howling suppression device, characterized in that: The device comprises: A sequence processing module, used for acquiring an acoustic signal frame sequence and an echo signal frame sequence, and processing the acoustic signal frame sequence and the echo signal frame sequence to obtain an acoustic periodic signal and an echo periodic signal; A sub-signal matching module, used for dividing the acoustic periodic signal and the echo periodic signal respectively, and matching the sub-signals in the acoustic periodic signal and the echo periodic signal to obtain a sub-signal matching pair; a howling suppression module, configured to perform howling detection on the sub-signal matching pair to obtain a howling sub-signal matching pair, and perform howling suppression on the howling sub-signal matching pair to obtain a howling suppression sub-signal; The output module is used to perform howling suppression on the sound periodic signal and the echo periodic signal according to the howling suppression sub-signal.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

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

Images