Speech enhancement method, apparatus, device, medium, and program product

Abstract

The application provides a speech enhancement method, device, equipment, medium and program product. The speech enhancement method comprises the following steps: performing speech enhancement on a microphone receiving signal to obtain a first enhanced signal; performing speech enhancement on a signal in a low frequency band of the first enhanced signal to obtain a second enhanced signal; determining a target enhanced signal based on the first enhanced signal and the second enhanced signal; and performing sound amplification on the target enhanced signal to obtain an amplified signal. The application can reduce the probability of howling in a sound amplification environment.

Description

Technical Field

[0001] This application relates to the field of speech processing technology, and in particular to a speech enhancement method, apparatus, device, medium, and program product. Background Technology

[0002] Amplifiers are primarily used to amplify a speaker's voice. However, due to differences in speaker volume, environment, and equipment, signals played through amplifiers are often affected by ambient noise, room reverberation, and feedback from the amplifier, resulting in poor voice quality and a tendency for feedback. Currently, feedback is relatively common in amplified environments. Summary of the Invention

[0003] This application provides a speech enhancement method, apparatus, device, medium, and program product to reduce the probability of feedback in a loudspeaker environment.

[0004] According to a first aspect of the embodiments of this application, a speech enhancement method is provided, comprising:

[0005] The microphone received signal is enhanced to obtain the first enhanced signal;

[0006] Speech enhancement is performed on the low-frequency band of the first enhanced signal to obtain the second enhanced signal;

[0007] The target enhancement signal is determined based on the first enhancement signal and the second enhancement signal;

[0008] The target enhancement signal is amplified to obtain the amplified signal.

[0009] Optionally, the step of enhancing the microphone received signal to obtain a first enhanced signal includes:

[0010] Perform a Fourier transform on the microphone received signal to obtain its frequency domain characteristics;

[0011] Based on the frequency domain characteristics, power spectrum characteristics and complex spectrum characteristics are obtained;

[0012] Based on the power spectrum features and the complex spectrum features, predict the time-frequency domain mask;

[0013] Based on the frequency domain features and the time-frequency domain mask, a first enhanced signal is obtained.

[0014] Optionally, the step of performing speech enhancement on the low-frequency band signal in the first enhanced signal to obtain the second enhanced signal includes:

[0015] Based on the power spectrum characteristics and the complex spectrum characteristics, the estimated complex coefficients of the filter corresponding to the low frequency band are obtained;

[0016] The second enhanced signal is obtained by estimating the complex coefficients of the filter corresponding to the low-frequency band and the signal in the low-frequency band of the first enhanced signal.

[0017] Optionally, a filter order, a target frame, and a frequency correspond to a filter estimation complex coefficient; in the frequency domain features, the acquisition time of the target frame is greater than or equal to the acquisition time of the current frame;

[0018] The step of estimating complex coefficients based on the filter corresponding to the low-frequency band and obtaining the second enhanced signal from the low-frequency band signal in the first enhanced signal includes:

[0019] Based on the filter estimation complex coefficients of each target frame in the filter estimation complex coefficients corresponding to any filter order and the low-frequency band, and the signal in the low-frequency band of the first enhancement signal, the intermediate enhancement signal corresponding to the arbitrary filter order and the low-frequency band is determined.

[0020] A second enhanced signal is obtained based on each filter order and the intermediate enhanced signal corresponding to the low-frequency band.

[0021] Optionally, determining the target enhancement signal based on the first enhancement signal and the second enhancement signal includes:

[0022] Based on the power spectrum characteristics and the complex spectrum characteristics, the weighting coefficients are determined;

[0023] The third enhancement signal is determined based on the first enhancement signal, the second enhancement signal, and the weighting coefficients;

[0024] The third enhanced signal is subjected to an inverse Fourier transform to obtain the fourth enhanced signal;

[0025] Based on the fourth enhancement signal, the target enhancement signal is determined.

[0026] Optionally, determining the target enhancement signal based on the fourth enhancement signal includes:

[0027] The feedback signal in the fourth enhanced signal is subjected to frequency shifting, phase modulation, and half-wave rectification to remove the correlation between the feedback signal and the target speech signal in the fourth enhanced signal, thereby obtaining the fifth enhanced signal.

[0028] Based on the fifth enhancement signal, the target enhancement signal is determined;

[0029] The feedback signal refers to the signal played back from the speaker to the microphone; the target voice signal refers to the user's voice signal received by the microphone.

[0030] Optionally, determining the target enhancement signal based on the fifth enhancement signal includes:

[0031] The speaker output signal is determined based on the fifth enhancement signal;

[0032] The dynamic range of the speaker output signal is mapped to a preset dynamic range to obtain the adjusted speaker output signal;

[0033] The target enhancement signal is determined based on the adjusted speaker output signal.

[0034] According to a second aspect of the embodiments of this application, a voice enhancement device is provided, comprising:

[0035] The first enhancement unit is used to enhance the voice signal received by the microphone to obtain the first enhanced signal;

[0036] The second enhancement unit is used to perform speech enhancement on the low-frequency band signal in the first enhancement signal to obtain the second enhancement signal;

[0037] A processing unit is configured to determine a target enhancement signal based on the first enhancement signal and the second enhancement signal;

[0038] A loudspeaker unit is used to amplify the target enhancement signal to obtain an amplified signal.

[0039] According to a third aspect of the embodiments of this application, an electronic device is provided, including a memory and a processor;

[0040] The memory is connected to the processor and is used to store programs;

[0041] The processor is used to implement the speech enhancement method as described in the first aspect by running a program in the memory.

[0042] According to a fourth aspect of the embodiments of this application, a storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the speech enhancement method as described in the first aspect.

[0043] According to a fifth aspect of the embodiments of this application, a computer program product is provided, including computer program instructions that, when executed by a processor, cause the processor to perform the speech enhancement method as described in the first aspect.

[0044] In this application, the microphone received signal is enhanced to obtain a first enhanced signal. Then, the low-frequency band of the first enhanced signal is further enhanced to obtain a second enhanced signal. Based on the first and second enhanced signals, a target enhanced signal is determined, and the target enhanced signal is amplified to obtain amplified signal. The microphone received signal undergoes two voice enhancements. The first enhancement is a full-band enhancement, which reduces the probability of feedback. The second enhancement focuses on the low-frequency band of the first enhanced signal. Since low-frequency signals have stronger energy and are more prone to feedback, enhancing the low-frequency band of the first enhanced signal to obtain the second enhanced signal further reduces the probability of feedback. Based on the first and second enhanced signals, the target enhanced signal is determined and amplified to obtain amplified signal, further reducing the probability of feedback in the amplified environment. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating a speech enhancement method provided in an embodiment of this application;

[0047] Figure 2 This is a schematic diagram illustrating the source of a microphone receiving signal in a loudspeaker scenario provided in this application embodiment;

[0048] Figure 3 This is a flowchart illustrating step 101 provided in an embodiment of this application;

[0049] Figure 4 This is a flowchart illustrating step 102 provided in an embodiment of this application;

[0050] Figure 5 This is a flowchart illustrating step 402 provided in an embodiment of this application;

[0051] Figure 6 This is a flowchart illustrating step 103 provided in an embodiment of this application;

[0052] Figure 7 This is a flowchart illustrating step 604 provided in an embodiment of this application;

[0053] Figure 8This is a flowchart illustrating step 702 provided in an embodiment of this application;

[0054] Figure 9 This is a schematic diagram of the structure of an exemplary adaptive feedback suppression model provided in the embodiments of this application;

[0055] Figure 10 This is a schematic diagram of the structure of a speech enhancement device provided in the embodiments of this application;

[0056] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0057] Megaphones are primarily used to amplify a speaker's voice. However, due to differences in speaker volume, environment, and equipment, signals played through megaphones are often affected by ambient noise, room reverberation, and feedback from the megaphone, resulting in degraded voice quality and a tendency for feedback. Furthermore, due to variations in hardware algorithms, amplification latency is often significant, leading to a poor listening experience. Currently, feedback is quite common in amplified environments.

[0058] Existing solutions for suppressing speaker feedback primarily focus on hardware, such as using directional microphones and reducing the maximum gain of the speaker to decrease the probability of feedback. Other approaches involve adding filters to attenuate the low-frequency energy of the output speech or reducing the maximum output gain of the speaker to balance amplification and feedback reduction. However, these methods are ineffective against feedback that has already occurred in the system and only provide a passive defense. Furthermore, due to the portability of loudspeakers, existing noise reduction algorithms for loudspeakers suffer from slow convergence and significant latency.

[0059] To reduce the probability of feedback in a loudspeaker environment, this application provides a speech enhancement method, apparatus, device, medium, and program product.

[0060] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] Exemplary Implementation Environment

[0062] The voice enhancement method according to embodiments of this application can be executed by electronic devices such as terminal devices or servers. The terminal device can be a user device, mobile device, computing device, wearable device, etc., and the server can be an independent physical server, a server cluster consisting of multiple physical servers, or a cloud server capable of cloud computing. This method can be implemented by a processor calling computer-readable program instructions stored in memory.

[0063] Exemplary methods

[0064] Please see Figure 1 In one exemplary embodiment, a speech enhancement method is provided. For example... Figure 1 As shown, the main steps of the speech enhancement method include:

[0065] Step 101: Perform voice enhancement on the microphone received signal to obtain the first enhanced signal.

[0066] In some embodiments, such as Figure 2 The diagram shown illustrates the source of the microphone signal received in a loudspeaker scenario. Figure 2 In this diagram, the microphone received signal is denoted by y(t), the target speech signal by s(t), the feedback signal by d(t), the background noise by n(t), the room reverberation by r(t), and the speaker playback reference signal by x(t). G is the speaker amplification gain, and h(t) represents the attenuation during the propagation from the speaker to the microphone. Specifically, the target speech signal s(t) refers to the user's speech signal received by the microphone, i.e., the speaker's direct speech signal; the feedback signal d(t) refers to the signal played back from the speaker to the microphone, i.e., the speaker's direct signal; and the room reverberation r(t) refers to the signal from the speaker's speech and the speaker signal that propagates to the microphone after reflection through the room. In a loudspeaker setup, the microphone received signal y(t) can be mainly divided into three components: (1) the target speech signal s(t); (2) the feedback signal d(t); and (3) the background noise n(t) and the room reverberation r(t).

[0067] In the exemplary embodiment, the specific formula for the feedback signal d(t) is as follows:

[0068] d(t) = NL(x(t)) * h(t)

[0069] Where NL() represents the nonlinear transformation of the loudspeaker, x(t) represents the reference signal played by the loudspeaker, h(t) represents the attenuation during the propagation process from the loudspeaker to the microphone, and * represents linear convolution.

[0070] Theoretically, without any other processing, the reference signal x(t) played by the speaker is obtained by delaying and amplifying the signal y(t) received by the microphone, and this signal will be repeatedly picked up by the microphone. The formula for the microphone received signal y(t) at any time t is as follows:

[0071]

[0072] Where s(t) represents the target speech signal, n(t) represents the background noise, r(t) represents the room reverberation, Δt represents the system transmission delay from the microphone to the speaker, NL() represents the speaker nonlinear transformation, G is the speaker amplification gain, I is the total time for feedback signal reamplification, and h(t) represents the attenuation during the propagation process from the speaker to the microphone.

[0073] As can be seen from the above formula, repeated pickup of the feedback signal will form positive feedback, which will lead to cyclic amplification at certain frequencies, resulting in howling.

[0074] In some embodiments, such as Figure 3 As shown, step 101 includes:

[0075] Step 301: Perform a Fourier transform on the microphone received signal to obtain frequency domain characteristics.

[0076] In some embodiments, prior to step 301, the voice enhancement method further includes: filtering out signals below 150Hz from the microphone received signal using a high-pass filter.

[0077] Among them, signals below 150Hz are mainly background noise and have no impact on the target speech signal.

[0078] In the exemplary embodiment, the Fourier transform can be an STFT (Short-Time Fourier Transform) or other types of Fourier transforms, and this application does not limit it.

[0079] In the exemplary embodiment, the microphone received signal is represented by y(t), and the frequency domain feature is represented by X(k,f), where k represents the frame number and f represents the frequency point.

[0080] In the exemplary embodiment, the length of the frequency domain feature of each frame is 256 points, that is, the synthesis window length is 256 points, the difference between the starting points of the frequency domain features of adjacent frames is 80 points, that is, the frame shift is 80 points, the sampling rate is 16,000 times per second, and the processing delay is (256+80) / 16,000 = 21ms. This meets the low latency requirements of the loudspeaker system, reduces the link latency, and ensures that there is no significant time difference between the speaker's direct signal and the loudspeaker output signal. The human ear cannot hear the time delay between the speaker's direct sound and the speaker's sound, thus avoiding the "echo" problem in loudspeaker systems.

[0081] Step 302: Based on the frequency domain characteristics, obtain the power spectrum characteristics and complex spectrum characteristics.

[0082] In an exemplary embodiment, the power spectrum feature may be a log power spectrum (LPS) feature or other types of power spectrum features, and this application does not limit it to these.

[0083] In the exemplary embodiment, the frequency domain characteristics are represented by X(k,f); the power spectrum characteristics are represented by X0. LPS (k,f) represents the complex spectrum characteristics, denoted by X. DF (k,f′) represents: k represents the number of frames; f represents the frequency point; f′ represents the frequency point of the low frequency band, and f′=f / 2.

[0084] Power spectral characteristics X LPS The formula for calculating (k,f) is as follows:

[0085] X LPS (k,f)=20log 10 (Re(X(k,f)) 2 +Im(X(k,f)) 2 )

[0086] Where Re represents the real part of the complex number, and Im represents the imaginary part of the complex number.

[0087] Complex spectral characteristics X DF The formula for calculating (k,f′) is as follows:

[0088] X DF (k,f′)=(Re(X(k,f / 2)),Im(X(k,f / 2)))

[0089] Where Re represents the real part of the complex number, Im represents the imaginary part of the complex number, f′ represents the frequency point of the low-frequency band, and f′=f / 2.

[0090] Step 303: Predict the time-frequency domain mask based on power spectrum features and complex spectrum features.

[0091] In an exemplary embodiment, the power spectrum characteristics are represented by X. LPS(k,f) represents the complex spectrum characteristics, denoted by X. DF (k,f′) represents the time-frequency domain mask, which is represented by G(k,f); k represents the number of frames (time domain dimension); and f represents the frequency point (frequency domain dimension).

[0092] In an exemplary embodiment, step 303 may include: inputting power spectrum features and complex spectrum features into an encoding / decoding network to obtain a time-frequency domain mask output by the encoding / decoding network. Specifically, this may involve inputting power spectrum features and complex spectrum features into an encoder to obtain the encoder's output features; and inputting the encoder's output features into a decoder to obtain a time-frequency domain mask output by the decoder.

[0093] Step 304: Based on the frequency domain features and the time-frequency domain mask, obtain the first enhanced signal.

[0094] In the exemplary embodiment, the frequency domain feature is represented by X(k,f); the time-frequency domain mask is represented by G(k,f); and the first enhancement signal is represented by Y. G (k,f) represents the expression.

[0095] First enhanced signal Y G The specific formula for (k,f) is as follows:

[0096] Y G (k,f)=X(k,f)·G(k,f)

[0097] A Fourier transform is performed on the microphone received signal to obtain frequency domain features. Based on these features, power spectrum and complex spectrum features are obtained. A time-frequency domain mask is predicted based on these features. Finally, the first enhanced signal is obtained based on the frequency domain features and the time-frequency domain mask. The estimated real-valued target signal mask, i.e., the time-frequency domain mask, is applied to the original microphone signal to obtain the first-stage enhanced spectral envelope, i.e., the first enhanced signal. This estimation and enhancement of the full-band signal, for a loudspeaker system, primarily suppresses noise and reverberation components in the microphone received signal, thereby improving the signal-to-noise ratio (SNR) of the microphone received signal.

[0098] In some other embodiments, step 101 includes: performing voice enhancement on the microphone received signal through a frequency domain filter to obtain a first enhanced signal.

[0099] Step 102: Perform speech enhancement on the low-frequency band of the first enhanced signal to obtain the second enhanced signal.

[0100] In some embodiments, the selection of the low-frequency band signal in the first enhancement signal includes, but is not limited to, the following methods:

[0101] Method 1

[0102] The frequency of the low-frequency band is less than or equal to the mid-frequency of the first enhancement signal.

[0103] In an exemplary embodiment, the intermediate frequency of the first enhanced signal can refer to a frequency located in the middle of the frequency band of the first enhanced signal. For example, the intermediate frequency of the first enhanced signal can refer to half of the maximum frequency of the first enhanced signal; the intermediate frequency of the first enhanced signal can also refer to the average of the maximum and minimum frequencies of the first enhanced signal; the intermediate frequency of the first enhanced signal can also be other frequencies, and this application does not limit this. In this application, the intermediate frequency of the first enhanced signal being half of the maximum frequency of the first enhanced signal is used as an example for explanation.

[0104] Method 2

[0105] The frequency of the low-frequency band is less than or equal to the preset ratio frequency of the first enhanced signal, wherein the preset ratio frequency is equal to the frequency obtained by multiplying the maximum frequency of the first enhanced signal by the preset ratio.

[0106] Method 3

[0107] The low-frequency band has a frequency less than or equal to a preset frequency, where the preset frequency value can be set by the user.

[0108] The microphone received signal underwent two voice enhancement processes. The first enhancement applied full-band voice enhancement to reduce the probability of feedback. The second enhancement focused on the low-frequency band of the first enhanced signal, which is more prone to feedback due to its higher energy. This second enhancement further reduced the probability of feedback. Based on these two enhancements, a target enhancement signal was determined and amplified to further reduce feedback in the amplified environment. This method significantly improves the stability of the amplification system and reduces feedback by utilizing existing equipment without incurring additional hardware costs.

[0109] In some embodiments, step 102 includes: performing speech enhancement on the low-frequency band signal in the first enhancement signal using a frequency domain filter to obtain a second enhancement signal.

[0110] In other embodiments, such as Figure 4 As shown, step 102 includes:

[0111] Step 401: Based on the power spectrum characteristics and complex spectrum characteristics, obtain the estimated complex coefficients of the filter corresponding to the low frequency band.

[0112] In an exemplary embodiment, step 401 may include: inputting power spectrum features and complex spectrum features into the encoder to obtain the encoder's output features; and inputting the encoder's output features into the DeepFilterNet network to obtain the filter estimation complex coefficients corresponding to the low-frequency band.

[0113] DeepFilterNet is a low-complexity speech enhancement framework designed specifically for full-band audio (48kHz). Based on deep filtering technology, it processes audio signals through a neural network model to achieve efficient noise suppression.

[0114] Step 402: Based on the filter estimation of the low-frequency band and the signal in the low-frequency band of the first enhanced signal, obtain the second enhanced signal.

[0115] Based on power spectrum and complex spectrum characteristics, the estimated complex coefficients of the filter corresponding to the low-frequency band are obtained. Based on these estimated complex coefficients and the low-frequency signal in the first enhanced signal, a second enhanced signal is obtained. By estimating the complex target signal mask, i.e., the estimated complex coefficients of the filter, residual reverberation and feedback components in the first enhanced signal are further eliminated, residual speech frame interference signals in the first enhanced signal are suppressed, and the speech periodic components are further enhanced, actively enhancing the target speech components. Furthermore, since the signal energy in the low-frequency band is stronger, it is more prone to feedback howling. Obtaining the second enhanced signal based on the estimated complex coefficients of the filter corresponding to the low-frequency band and the low-frequency signal in the first enhanced signal can further reduce the probability of howling.

[0116] In some embodiments, a filter order, a target frame, and a frequency correspond to a filter estimation complex coefficient; in the frequency domain features, the acquisition time of the target frame is greater than or equal to the acquisition time of the current frame.

[0117] In an exemplary embodiment, the filter estimates the complex coefficients using C. i (k+j,f′) represents the filter order i, the frame number k, the current frame k, the target frame j to the right of the current frame (k+j), and the frequency point f′.

[0118] In some embodiments, such as Figure 5 As shown, step 402 includes:

[0119] Step 501: Based on the filter estimation complex coefficients of each target frame in the filter estimation complex coefficients corresponding to any filter order and low frequency band, and the signal in the low frequency band of the first enhancement signal, determine the intermediate enhancement signal corresponding to any filter order and low frequency band.

[0120] In the exemplary embodiment, the intermediate enhancement signal corresponding to the i-th filter order and the low-frequency band is represented by Y. i (k,f′) represents the intermediate enhancement signal Y. i The formula for calculating (k,f′) is as follows:

[0121]

[0122] Where i refers to the i-th filter order, k represents the frame number, the k-th frame is the current frame, (k+j) refers to the j-th target frame to the right of the current frame, f′ represents the frequency point of the low-frequency band, and the complex coefficients of the filter estimation are represented by C. i (k+j,f′) represents that l refers to the total number of frames to the right of the k-th frame, and Y G (k―i,f′) refers to the first enhanced signal corresponding to the i-th filter order and the low-frequency band.

[0123] Step 502: Obtain the second enhanced signal based on each filter order and the intermediate enhanced signal corresponding to the low frequency band.

[0124] In an exemplary embodiment, the second enhancement signal uses Y DF′ (k,f′) represents the intermediate enhancement signal corresponding to the i-th filter order and the low-frequency band, denoted by Y. i (k,f′) represents the second enhanced signal Y. DF′ The formula for calculating (k,f′) is as follows:

[0125]

[0126] Where N refers to the total filtering order.

[0127] Therefore, the second enhanced signal Y DF′ The formula for calculating (k,f′) is as follows:

[0128]

[0129] The first stage involves estimating and enhancing the full-band signal to obtain the first enhanced signal Y. G (k,f), the second stage enhances the low-frequency band of the signal based on the first stage. Since the low-frequency signal has strong energy, it is more likely to generate feedback howling. Using a complex filter can effectively recover the periodic components of the original signal and reduce the probability of system howling. By estimating the complex target signal mask, that is, the filter estimates the complex coefficients, and by filtering adjacent frames, the reverberation and feedback components remaining in the first enhanced signal are further eliminated, the speech frame interference signals remaining in the first enhanced signal are suppressed, the periodic components of the speech are further enhanced, and the target speech components are actively enhanced.

[0130] In some other embodiments, step 402 includes: directly applying the filter estimate complex coefficients corresponding to the low-frequency band to the low-frequency band signal in the first enhanced signal to obtain the second enhanced signal.

[0131] Step 103: Determine the target enhancement signal based on the first enhancement signal and the second enhancement signal.

[0132] In some embodiments, step 103 includes: performing an inverse Fourier transform on the signal obtained by adding the first enhanced signal and the second enhanced signal to obtain the target enhanced signal.

[0133] In other embodiments, such as Figure 6 As shown, step 103 includes:

[0134] Step 601: Determine the weighting coefficients based on the power spectrum characteristics and complex spectrum characteristics.

[0135] In the exemplary embodiment, the weighting coefficient is represented by α(k), where k represents the number of frames.

[0136] In an exemplary embodiment, step 601 may include: inputting power spectrum features and complex spectrum features into the encoder to obtain the encoder's output features; and inputting the encoder's output features into the DeepFilterNet network to obtain weighting coefficients.

[0137] DeepFilterNet is a low-complexity speech enhancement framework designed specifically for full-band audio (48kHz). Based on deep filtering technology, it processes audio signals through a neural network model to achieve efficient noise suppression.

[0138] Step 602: Determine the third enhanced signal based on the first enhanced signal, the second enhanced signal, and the weighting coefficients.

[0139] In an exemplary embodiment, the first enhancement signal is represented by Y. G (k,f) represents the second enhancement signal, denoted by Y. DF′

[0140] (k,f′) represents the weighting coefficient α(k), k represents the frame number, f represents the frequency point, f′ represents the frequency point of the low-frequency band, and the third enhancement signal is represented by Y. DF (k,f) represents the third enhanced signal Y. DF The formula for calculating (k,f) is as follows:

[0141] Y DF (k,f)=α(k)·Y DF′ (k,f′)+(1―α(k))·Y G (k,f)

[0142] Step 603: Perform an inverse Fourier transform on the third enhanced signal to obtain the fourth enhanced signal.

[0143] In the exemplary embodiment, the inverse Fourier transform can be ISTFT (Inverse Short-Time Fourier Transform) or other types of inverse Fourier transform, and this application does not limit it to these.

[0144] Step 604: Determine the target enhancement signal based on the fourth enhancement signal.

[0145] Based on power spectrum characteristics and complex spectrum characteristics, weighting coefficients are determined. Based on the first enhanced signal, the second enhanced signal, and the weighting coefficients, the third enhanced signal is determined. The first enhanced signal is obtained by full-band speech enhancement of the microphone received signal. The second speech enhancement is performed on the low-frequency signal of the first enhanced signal after the first speech enhancement. Since the low-frequency signal has stronger energy, it is more likely to generate feedback howling. Performing speech enhancement on the low-frequency signal of the first enhanced signal to obtain the second enhanced signal can further reduce the probability of howling. In the process of determining the third enhanced signal, both the first and second enhanced signals are considered. Moreover, the weighting coefficients determined by power spectrum characteristics and complex spectrum characteristics are used to weight the first and second enhanced signals to obtain the third enhanced signal. This can preserve the target speech signal and suppress interference signals, thereby reducing the probability of howling.

[0146] In some embodiments, step 604 includes: directly determining the fourth enhancement signal as the target enhancement signal.

[0147] In other embodiments, such as Figure 7 As shown, step 604 includes:

[0148] Step 701: Perform frequency shifting, phase modulation, and half-wave rectification on the feedback signal in the fourth enhancement signal to remove the correlation between the feedback signal and the target speech signal in the fourth enhancement signal, and obtain the fifth enhancement signal.

[0149] The feedback signal refers to the signal played back from the speaker to the microphone; the target voice signal refers to the user's voice signal received by the microphone.

[0150] In the exemplary embodiment, frequency shifting and phase modulation refers to PMFS (Phase Modulation and Frequency Shifting), which involves phase modulation (PM) and frequency shifting (FS).

[0151] In the exemplary embodiment, half-wave rectification refers to Half-wave Rectification, where Rectification is usually represented by RECT.

[0152] In the exemplary embodiment, y(t) = s(t) + n(t) + r(t) + d(t), where the microphone received signal is represented by y(t), the target speech signal is represented by s(t), the feedback signal is represented by d(t), the background noise is represented by n(t), and the room reverberation is represented by r(t). The feedback signal in the fourth enhancement signal can be frequency-shifted, phase-modulated, and half-wave rectified to remove the correlation between the feedback signal and the target speech signal in the fourth enhancement signal, that is, to remove the correlation between d(t) and s(t). After removing the correlation, the formula for the microphone received signal y(t) is as follows:

[0153] y(t) = s(t) + z(t)

[0154] Where z(t) refers to noise interference independent of the target speech signal s(t), and the formula for z(t) is as follows:

[0155] z(t) = n(t) + r(t) + d(t)

[0156] In z(t), d(t) and s(t) have been decorrelated and are no longer correlated.

[0157] Step 702: Determine the target enhancement signal based on the fifth enhancement signal.

[0158] Because the feedback signal and the target speech signal in the loudspeaker system are highly correlated, the fourth enhancement signal obtained through nonlinear processing may contain distortion of the target speech signal. Therefore, frequency shifting, phase modulation and half-wave rectification are performed on the feedback signal in the fourth enhancement signal to remove the correlation between the feedback signal and the target speech signal and obtain the fifth enhancement signal. Based on the fifth enhancement signal, the target enhancement signal is determined, which can reduce the correlation between the feedback signal and the target speech signal and reduce the occurrence of distortion of the target speech signal.

[0159] In some embodiments, step 702 includes: directly determining the fifth enhancement signal as the target enhancement signal.

[0160] In other embodiments, such as Figure 8 As shown, step 702 includes:

[0161] Step 801: Determine the speaker output signal based on the fifth enhancement signal.

[0162] Step 802: Map the dynamic range of the speaker output signal to a preset dynamic range to obtain the adjusted speaker output signal.

[0163] In an exemplary embodiment, mapping the dynamic range of the speaker output signal to a preset dynamic range may refer to DRC (Dynamic Range Control).

[0164] Step 803: Determine the target enhancement signal based on the adjusted speaker output signal.

[0165] Based on the fifth enhancement signal, the speaker output signal is determined, and the dynamic range of the speaker output signal is mapped to the preset dynamic range to obtain the adjusted speaker output signal. Based on the adjusted speaker output signal, the target enhancement signal is determined, which can avoid howling caused by an excessively large speaker reference signal and reduce the probability of howling.

[0166] In some embodiments, step 803 includes: determining a sixth enhancement signal based on the adjusted speaker output signal; and processing the sixth enhancement signal through an equalizer to obtain a target enhancement signal.

[0167] Among them, the equalizer (EQ) can suppress some frequency bands that are prone to feedback and compensate for speech distortion.

[0168] Step 104: Amplify the target enhancement signal to obtain the amplified signal.

[0169] In an exemplary embodiment, step 104 may include: amplifying the target enhancement signal through a loudspeaker to obtain the amplified signal.

[0170] In the exemplary embodiment, the microphones in commonly used loudspeakers are mainly of two types: wired directional microphones and wireless omnidirectional microphones. Wired directional microphones have a lower probability of feedback because they pick up sound in a directional manner and are relatively farther away from the speaker. Therefore, the voice enhancement method in this application is mainly aimed at wireless omnidirectional microphones.

[0171] In an exemplary embodiment, such as Figure 9 The diagram shown is a schematic representation of an exemplary adaptive feedback suppression model. Figure 9 In this model, the Adaptive Feedback Suppression (AFS) consists of a first-stage network and a second-stage network. The first-stage network includes an encoder and a decoder, while the second-stage network includes DeepFilterNet. DeepFilterNet is a low-complexity speech enhancement framework designed specifically for full-band audio (48kHz). Based on deep filtering technology, it processes audio signals through a neural network model to achieve efficient noise suppression. Figure 9In this context, Conv refers to convolution; GGRU refers to Group Gated Recurrent Unit; and Linear refers to linear processing.

[0172] according to Figure 9 An exemplary speech enhancement method derived from the exemplary adaptive feedback suppression model in the example mainly includes: performing a Fourier transform on the microphone received signal to obtain frequency domain features; obtaining power spectrum features and complex spectrum features based on the frequency domain features; inputting the power spectrum features and complex spectrum features into an encoder to obtain the encoder's output features; inputting the encoder's output features into a decoder to obtain the decoder's output time-frequency domain mask; obtaining a first enhanced signal based on the frequency domain features and the time-frequency domain mask; inputting the encoder's output features into DeepFilterNet to obtain the filter estimation complex coefficients and weighting coefficients corresponding to the low-frequency band; obtaining a second enhanced signal based on the filter estimation complex coefficients corresponding to the low-frequency band and the low-frequency signal in the first enhanced signal; determining a third enhanced signal based on the first enhanced signal, the second enhanced signal, and the weighting coefficients; performing an inverse Fourier transform on the third enhanced signal to obtain a fourth enhanced signal; determining a target enhanced signal based on the fourth enhanced signal; and amplifying the target enhanced signal to obtain the amplified signal.

[0173] Figure 9 The exemplary adaptive feedback suppression model in this application is a multi-stage noise reduction and compression model designed with reference to DeepFilterNet. It can improve the model's suppression effect on multi-component (noise, reverberation, and feedback) signals, and Figure 9 The exemplary adaptive feedback suppression model in this paper has fewer parameters, faster computation speed, and similar noise reduction performance to the traditional DeepFilterNet compared to the traditional DeepFilterNet. As shown in Table 1, Figure 9 The table shows the network parameter count and computational cost of the exemplary adaptive feedback suppression model and the traditional DeepFilterNet under the same comparable conditions, as well as the denoising effect of the models at different signal-to-noise ratios.

[0174] Table 1

[0175]

[0176] In Table 1, SNR (Signal-to-Noise Ratio), SDR (Signal-to-Distortion Ratio), and PESQ (Perceptual Evaluation of Speech Quality) are used. For both SDR and PESQ, a higher value is better.

[0177] As can be seen from Table 1, Figure 9 The exemplary adaptive feedback suppression model in this paper has fewer parameters, faster computation speed, and similar noise reduction effect to the traditional DeepFilterNet compared to the traditional DeepFilterNet.

[0178] In summary, this application involves two voice enhancement processes: first, full-band voice enhancement of the microphone received signal to obtain a first enhanced signal; second, voice enhancement of the low-frequency band of the first enhanced signal to obtain a second enhanced signal; third, determination of a target enhanced signal based on the first and second enhanced signals; and fourth, amplification of the target enhanced signal to obtain the amplified signal. The microphone received signal undergoes two voice enhancement processes. The first enhancement is a full-band enhancement, which reduces the probability of feedback. The second enhancement focuses on the low-frequency band of the first enhanced signal, which, due to its higher energy, is more prone to feedback. Amplifying the low-frequency band of the first enhanced signal to obtain the second enhanced signal further reduces the probability of feedback. Finally, determination of the target enhanced signal based on the first and second enhanced signals, followed by amplification of the target enhanced signal to obtain the amplified signal, further reduces the probability of feedback in the amplified environment.

[0179] Exemplary device

[0180] Accordingly, embodiments of this application also provide a voice enhancement device, such as... Figure 10 As shown, the voice enhancement device includes:

[0181] The first enhancement unit 1001 is used to enhance the microphone received signal to obtain a first enhanced signal;

[0182] The second enhancement unit 1002 is used to perform speech enhancement on the low-frequency band signal in the first enhancement signal to obtain the second enhancement signal;

[0183] Processing unit 1003 is configured to determine a target enhancement signal based on the first enhancement signal and the second enhancement signal;

[0184] The amplification unit 1004 is used to amplify the target enhancement signal to obtain the amplified signal.

[0185] Optionally, the first enhancement unit 1001 is specifically used for:

[0186] Perform a Fourier transform on the microphone received signal to obtain its frequency domain characteristics;

[0187] Based on the frequency domain characteristics, power spectrum characteristics and complex spectrum characteristics are obtained;

[0188] Based on the power spectrum features and the complex spectrum features, predict the time-frequency domain mask;

[0189] Based on the frequency domain features and the time-frequency domain mask, a first enhanced signal is obtained.

[0190] Optionally, the second enhancement unit 1002 includes:

[0191] The filter estimation subunit is used to obtain the filter estimation complex coefficients corresponding to the low-frequency band based on the power spectrum characteristics and the complex spectrum characteristics.

[0192] The first processing subunit is used to estimate the complex coefficients based on the filter corresponding to the low-frequency band and the signal in the low-frequency band of the first enhanced signal to obtain the second enhanced signal.

[0193] Optionally, a filter order, a target frame, and a frequency correspond to a filter estimation complex coefficient; in the frequency domain features, the acquisition time of the target frame is greater than or equal to the acquisition time of the current frame;

[0194] The first processing subunit is specifically used for:

[0195] Based on the filter estimation complex coefficients of each target frame in the filter estimation complex coefficients corresponding to any filter order and the low-frequency band, and the signal in the low-frequency band of the first enhancement signal, the intermediate enhancement signal corresponding to the arbitrary filter order and the low-frequency band is determined.

[0196] A second enhanced signal is obtained based on each filter order and the intermediate enhanced signal corresponding to the low-frequency band.

[0197] Optionally, the processing unit 1003 includes:

[0198] The weighting coefficient determination subunit is used to determine the weighting coefficients based on the power spectrum characteristics and the complex spectrum characteristics;

[0199] The second processing subunit is used to determine the third enhancement signal based on the first enhancement signal, the second enhancement signal, and the weighting coefficients;

[0200] The inverse transform subunit is used to perform an inverse Fourier transform on the third enhanced signal to obtain the fourth enhanced signal;

[0201] The third processing subunit is used to determine the target enhancement signal based on the fourth enhancement signal.

[0202] Optionally, the third processing subunit is specifically used for:

[0203] The feedback signal in the fourth enhanced signal is subjected to frequency shifting, phase modulation, and half-wave rectification to remove the correlation between the feedback signal and the target speech signal in the fourth enhanced signal, thereby obtaining the fifth enhanced signal.

[0204] Based on the fifth enhancement signal, the target enhancement signal is determined;

[0205] The feedback signal refers to the signal played back from the speaker to the microphone; the target voice signal refers to the user's voice signal received by the microphone.

[0206] Optionally, the third processing subunit is specifically used for:

[0207] The speaker output signal is determined based on the fifth enhancement signal;

[0208] The dynamic range of the speaker output signal is mapped to a preset dynamic range to obtain the adjusted speaker output signal;

[0209] The target enhancement signal is determined based on the adjusted speaker output signal.

[0210] The speech enhancement device provided in this embodiment belongs to the same concept as the speech enhancement method provided in the above embodiments of this application. It can execute the speech enhancement method provided in any of the above embodiments of this application and has the corresponding functional modules and beneficial effects for executing the speech enhancement method. Technical details not described in detail in this embodiment can be found in the specific processing content of the speech enhancement method provided in the above embodiments of this application, and will not be repeated here.

[0211] The functions implemented by the first enhancement unit 1001, the second enhancement unit 1002, the processing unit 1003, and the amplification unit 1004 described above can be implemented by the same or different processors, and this application embodiment does not limit this.

[0212] It should be understood that the units in the above device can be implemented by a processor calling software. For example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of each unit in the device. The processor can be a general-purpose processor, such as a CPU or microprocessor, and the memory can be internal or external to the device. Alternatively, the units in the device can be implemented as hardware circuits. By designing the hardware circuits, some or all of the unit functions can be implemented. The hardware circuits can be understood as one or more processors. For example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units are implemented by designing the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a PLD, such as an FPGA, which can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files to implement the functions of some or all of the above units. All units in the above device can be implemented entirely by a processor calling software, entirely by hardware circuits, or partially by a processor calling software with the remaining parts implemented by hardware circuits.

[0213] In this application embodiment, a processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction reading and execution capabilities, such as a CPU, microprocessor, GPU, or DSP. In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented as an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the processor loading instructions to implement the functions of some or all of the above units. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, or DPU.

[0214] As can be seen, each unit in the above device can be one or more processors (or processing circuits) configured to implement the above methods, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0215] Furthermore, the units in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a System-on-Chip (SoC). The SoC may include at least one processor for implementing any of the above methods or implementing the functions of the units in the device. The at least one processor may be of different types, such as CPU and FPGA, CPU and artificial intelligence processor, CPU and GPU, etc.

[0216] Exemplary electronic devices

[0217] One embodiment of this application discloses an electronic device, see [link to relevant documentation] Figure 11 As shown, the device includes:

[0218] Memory 200 and processor 210;

[0219] The memory 200 is connected to the processor 210 and is used to store programs;

[0220] The processor 210 is configured to implement the speech enhancement method disclosed in any of the above embodiments by running the program stored in the memory 200.

[0221] Specifically, the aforementioned electronic device may also include: a bus, a communication interface 220, an input device 230, and an output device 240.

[0222] The processor 210, memory 200, communication interface 220, input device 230, and output device 240 are interconnected via a bus. Among them:

[0223] A bus can include a pathway for transmitting information between various components of a computer system.

[0224] The processor 210 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0225] Processor 210 may include a main processor, as well as a baseband chip, modem, etc.

[0226] The memory 200 stores a program that executes the technical solution of this invention, and may also store an operating system and other key business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory 200 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0227] Input device 230 may include a device for receiving user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor.

[0228] Output device 240 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0229] The communication interface 220 may include a device that uses any transceiver to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0230] The processor 210 executes the program stored in the memory 200 and calls other devices, which can be used to implement the various steps of any of the speech enhancement methods provided in the above embodiments of this application.

[0231] Exemplary computer program products and storage media

[0232] In addition to the methods and devices described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the speech enhancement methods according to various embodiments of this application as described in any of the above embodiments of this specification.

[0233] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0234] Furthermore, embodiments of this application may also be storage media storing a computer program, which is executed by a processor to perform the steps of the speech enhancement method according to various embodiments of this application described in any of the above embodiments of this specification, specifically implementing the following steps:

[0235] Step 101: Perform voice enhancement on the microphone received signal to obtain the first enhanced signal.

[0236] Step 102: Perform speech enhancement on the low-frequency band of the first enhanced signal to obtain the second enhanced signal.

[0237] Step 103: Determine the target enhancement signal based on the first enhancement signal and the second enhancement signal.

[0238] Step 104: Amplify the target enhancement signal to obtain the amplified signal.

[0239] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0240] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0241] The steps in the methods of the various embodiments of this application can be adjusted, merged, or deleted in order according to actual needs, and the technical features described in each embodiment can be replaced or combined.

[0242] The modules and sub-modules in the apparatus and terminal in the various embodiments of this application can be merged, divided, and deleted according to actual needs.

[0243] It should be understood that the disclosed terminals, devices, and methods can be implemented in other ways, given the several embodiments provided in this application. For example, the terminal embodiments described above are merely illustrative. For instance, the division of modules or sub-modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple sub-modules or modules may be combined or integrated into another module, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0244] The modules or submodules described as separate components may or may not be physically separate. The components that constitute a module or submodule may or may not be physical modules or submodules; that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0245] Furthermore, the functional modules or sub-modules in the various embodiments of this application can be integrated into one processing module, or each module or sub-module can exist physically separately, or two or more modules or sub-modules can be integrated into one module. The integrated modules or sub-modules described above can be implemented in hardware or in the form of software functional modules or sub-modules.

[0246] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0247] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software unit executed by a processor, or a combination of both. The software unit can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0248] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0249] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A speech enhancement method, characterized by, The method comprises: performing Fourier transform on the microphone received signal to obtain a frequency domain feature; obtaining a power spectrum feature and a complex spectrum feature based on the frequency domain feature; obtaining a first enhanced signal based on the power spectrum feature and the complex spectrum feature; performing speech enhancement on the signal of the low frequency band in the first enhanced signal to obtain a second enhanced signal; determining a weighting coefficient based on the power spectrum feature and the complex spectrum feature; determining a third enhanced signal based on the first enhanced signal, the second enhanced signal and the weighting coefficient; performing inverse Fourier transform on the third enhanced signal to obtain a fourth enhanced signal; determining a target enhanced signal based on the fourth enhanced signal; performing sound amplification on the target enhanced signal to obtain an amplified signal.

2. The speech enhancement method of claim 1, wherein, The method comprises: predicting a time-frequency domain mask based on the power spectrum feature and the complex spectrum feature; obtaining the first enhanced signal based on the frequency domain feature and the time-frequency domain mask.

3. The speech enhancement method of claim 1, wherein, The method comprises: obtaining a low-frequency band corresponding filter estimation complex coefficient based on the power spectrum feature and the complex spectrum feature; obtaining the second enhanced signal based on the low-frequency band corresponding filter estimation complex coefficient and the signal of the low frequency band in the first enhanced signal.

4. The speech enhancement method of claim 3, wherein, One filter estimation complex coefficient corresponds to one filter order, one target frame and one frequency; in the frequency domain feature, the collection time of the target frame is greater than or equal to the collection time of the current frame; The method comprises: determining an intermediate enhanced signal corresponding to the low frequency band and the arbitrary filter order based on the signal of the low frequency band in the first enhanced signal and the filter estimation complex coefficient corresponding to the low frequency band and each target frame of the filter estimation complex coefficient of the arbitrary filter order; obtaining the second enhanced signal based on each filter order and the intermediate enhanced signal corresponding to the low frequency band.

5. The speech enhancement method of claim 1, wherein, The method comprises: performing frequency shift and phase modulation and half-wave rectification on the feedback signal in the fourth enhanced signal to remove the correlation between the feedback signal and the target speech signal in the fourth enhanced signal to obtain a fifth enhanced signal; determining the target enhanced signal based on the fifth enhanced signal; wherein, the feedback signal refers to the signal played back to the microphone by the loudspeaker; the target speech signal refers to the user speech signal received by the microphone.

6. The speech enhancement method of claim 5, wherein, The method comprises: determining a loudspeaker output signal based on the fifth enhanced signal; mapping the dynamic range of the loudspeaker output signal to a preset dynamic range to obtain an adjusted loudspeaker output signal; determining the target enhanced signal based on the adjusted loudspeaker output signal.

7. A speech enhancement device, characterized by The method comprises: a first enhancement unit, configured to perform Fourier transform on the microphone received signal to obtain a frequency domain feature; obtain a power spectrum feature and a complex spectrum feature based on the frequency domain feature; obtain a first enhanced signal based on the power spectrum feature and the complex spectrum feature; a second enhancement unit, configured to perform voice enhancement on a signal in a low frequency band in the first enhanced signal to obtain a second enhanced signal; a processing unit, configured to determine a weighting coefficient based on the power spectrum feature and the complex spectrum feature, and determine a third enhanced signal based on the first enhanced signal, the second enhanced signal and the weighting coefficient; perform inverse Fourier transform on the third enhanced signal to obtain a fourth enhanced signal; determine a target enhanced signal based on the fourth enhanced signal; an amplification unit, configured to amplify the target enhanced signal to obtain an amplified signal.

8. An electronic device, comprising: comprising a memory and a processor; the memory is connected with the processor, and is configured to store a program; the processor is configured to realize the voice enhancement method in any one of claims 1 to 6 by running the program in the memory.

9. A storage medium, characterized by the storage medium has a computer program stored thereon, and the computer program, when being run by a processor, realizes the voice enhancement method in any one of claims 1 to 6.

10. A computer program product, characterised in that, comprising computer program instructions, and the computer program instructions, when being run by a processor, cause the processor to perform the voice enhancement method in any one of claims 1 to 6.

Images