Distributed optical fiber sensing speech enhancement method based on soil geology and environment information

Abstract

The invention relates to the technical field of distributed optical fiber acoustic sensing, in particular to a distributed optical fiber sensing speech enhancement method based on soil geology and environment information. Comprising the following steps: S1, collecting space-time sound wave data; s2, performing voice enhancement operation on the space-time sound wave data acquired in the step S1, specifically comprising the following steps: S21, performing feature extraction on the space-time sound wave data; s22, constructing a clue mask network based on the soil geological information and the soil environment information; s23, a complex spectrum convolution recurrent neural network is constructed, the complex spectrum convolution recurrent neural network comprises a clue mask network and a backbone network, the features extracted from the time-space sound wave data in the step S21 are input into the complex spectrum convolution recurrent neural network, and speech enhancement data of the time-space sound wave data are obtained; according to the invention, the external deployment environment information of the distributed optical fiber is fully utilized, and the performance of the distributed optical fiber sensing system is better improved.

Description

Technical Field

[0001] The invention relates to the technical field of distributed optical fiber acoustic sensing, and in particular to a distributed optical fiber sensing speech enhancement method based on soil geology and environmental information. Background Art

[0002] Distributed fiber optic acoustic wave sensing technology, as a new type of acoustic wave detection technology, uses a large area of ​​distributed optical fiber as a sensing unit, and detects the weak bending and twisting of the optical fiber caused by physical events in the surrounding environment, as well as the phase change of the backward Rayleigh scattered light in the optical fiber, to achieve real-time positioning and perception of external vibration sources or sound sources. Distributed fiber optic acoustic wave sensing technology has the characteristics of high sensitivity, high information abundance, and multi-dimensional signal characteristics. It has been widely used in perimeter security, water conservancy / oil and gas pipeline monitoring, earthquake monitoring and other fields. However, in the distributed acoustic sensing system, due to the complexity of voice events and the unknown nature of noise, the collected spatiotemporal data is seriously affected by random noise, and there are problems such as poor data quality and inability to extract effective information. Therefore, it is necessary to eliminate the influence of background noise from the collected spatiotemporal data (i.e., data enhancement / voice enhancement), improve data quality, and improve the performance of the distributed acoustic sensing system.

[0003] Traditional speech enhancement methods, including Wiener filtering, spectral subtraction, principal component analysis, etc., perform better in steady-state noise scenarios, but their performance is very limited in non-steady-state, random scenarios. In recent years, speech enhancement technology based on deep learning has used neural network training based on a large number of clean and noisy audio data sets to capture the nonlinear relationship between noise and clean signal features, improve data quality, and achieve speech enhancement.

[0004] Prior art CN115622626A discloses a distributed acoustic wave sensing voice information recognition system and method, which designs a distributed sensing optical fiber system, uses a sensing optical fiber to receive voice signals, uses a laser output unit to emit a narrowband laser signal, uses a circulator to detect the voice signal, and finally receives the backscattered signal through the acquisition unit and the acquisition circulator, and finally uses a convolutional recursive network unit to perform complex domain mapping on the scattered signal to reconstruct the voice signal. However, the above method focuses on the modular function realization of the distributed optical fiber system, does not fully utilize the external deployment environment information of the distributed optical fiber, and cannot achieve a better performance distributed optical fiber sensing system design.

[0005] Therefore, there is an urgent need to provide a distributed fiber optic sensing speech enhancement method based on soil geology and environmental information. Compared with the existing technology, it can make full use of the external deployment environment information of the distributed optical fiber and better improve the performance of the distributed fiber optic sensing system. Summary of the invention

[0006] The present invention solves the technical problems existing in the prior art and provides a distributed optical fiber sensing speech enhancement method based on soil geology and environmental information.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0008] The distributed optical fiber sensing voice enhancement method based on soil geology and environmental information includes the following steps:

[0009] S1. Collecting spatiotemporal acoustic wave data;

[0010] S2, performing a speech enhancement operation on the spatiotemporal sound wave data collected in step S1, specifically comprising the following steps:

[0011] S21, extracting features from spatiotemporal sound wave data;

[0012] S22, constructing a clue mask network based on soil geological information and soil environmental information;

[0013] S23. Construct a complex spectrum convolutional recurrent neural network, which includes a clue mask network and a backbone network. Input the features extracted from the spatiotemporal sound wave data in step S21 into the complex spectrum convolutional recurrent neural network to obtain speech enhancement data of the spatiotemporal sound wave data.

[0014] Furthermore, S21 specifically includes the following steps:

[0015] S211, performing short-time Fourier transform on the time domain signal of the space-time sound wave data to obtain the real part and the imaginary part of the short-time Fourier transform;

[0016] S212, setting a Mel filter bank, passing the real part and the imaginary part obtained in step S211 through the Mel filter bank respectively, to obtain a Mel short-time Fourier transform spectrum;

[0017] S213, taking the logarithm of the Mel short-time Fourier transform spectrum to obtain a logarithmic Mel time spectrum diagram;

[0018] S214. Perform first-order differences and second-order differences on the logarithmic Mel-spectrogram to obtain corresponding first-order differences of the logarithmic Mel spectrum and second-order differences of the logarithmic Mel spectrum, thereby obtaining features to be extracted from the spatiotemporal sound wave data. The features to be extracted from the spatiotemporal sound wave data include the logarithmic Mel-spectrogram, the first-order differences of the logarithmic Mel spectrum, and the second-order differences of the logarithmic Mel spectrum.

[0019] Furthermore, in step S213, the logarithmic Mel-time spectrum is obtained by the following formula:

[0020]

[0021] In the above formula, M(f,t) represents the value of the Mel spectrogram at frequency f and time t, |H j (f)| represents the response value of the jth Mel filter at frequency f, |X j (t)| represents the amplitude of the frequency domain signal corresponding to the jth Mel filter at time t, n represents the total number of Mel filters, and i is 1-n.

[0022] Further, S22 specifically includes the following steps:

[0023] S221, performing feature fusion of soil geological information and soil environmental information, the soil geological information includes soil depth and soil water content, the soil environmental information includes surface vegetation conditions, performing binary conversion on soil depth, soil water content, and surface vegetation conditions to obtain a one-hot encoding;

[0024] S222. Construct a clue mask network based on the one-hot encoding.

[0025] Furthermore, the specific method of constructing the clue mask network in step S222 is: setting a Transformer encoder, setting the one-hot encoding representation as l; inputting l into the Transformer encoder, and obtaining the output E of the Transformer encoder k , E k Integrate with l to perform multiplication query and get E k ′, E k 、E k ′ are projected to the decoder dimension N using a one-dimensional convolutional layer d 、N e , respectively get the projection code P k , P k ′, then, P k , P k ′ are respectively passed into the self-attention module and the cross-attention module, P k After being processed by the self-attention module and then passed through the feedforward neural network, it is input into the cross-attention module. The output of the cross-attention module is the target mask in the projection decoder space. The target mask passes through the feedforward neural network and the one-dimensional convolution layer in turn, and is projected back to the Transformer encoder dimension to obtain m k ′, and finally the clue mask vector is obtained, thus obtaining the clue mask network.

[0026] Furthermore, the cue mask vector is calculated as follows:

[0027] m k =m k ′+E k ′;

[0028] In the above formula, mk Represents the clue mask vector.

[0029] Furthermore, the backbone network includes an encoder, a long short-term memory network and a decoder, and the encoder and the decoder are quickly linked through the long short-term memory network. In step S21, the features extracted from the spatiotemporal sound wave data are input into the encoder, and the encoder processes and outputs them. The output of the encoder is input into the clue mask network to obtain a clue mask vector, and then the clue mask vector and the clue mask network are subjected to multiplication query integration. The result of the multiplication query integration is sequentially passed through the long short-term memory network and the decoder, and the output of the decoder is the speech enhancement data of the spatiotemporal sound wave data.

[0030] Furthermore, both the encoder and the decoder include four complex two-dimensional convolutions, complex batch normalization operations, and real PReLU activation functions. The specific expression of the complex two-dimensional convolution is:

[0031]

[0032] y[i] represents the output sequence, w[k] represents the convolution kernel, r represents the void ratio of the convolution kernel, K represents the size of the convolution kernel, k is 1-K, x[i-rk] represents the sequence after the input sequence is processed, and x[i] represents the input sequence.

[0033] Furthermore, the long short-term memory network processes the real part and the imaginary part of the input separately, and the processing process is specifically expressed by the following formula:

[0034] F rr =LSTM r (X r );

[0035] F ir =LSTM r (X i );

[0036] F ri =LSTM i (X r );

[0037] F ii =LSTM i (X i );

[0038] F out =(F rr -F ii )+j(F ir +F ri );

[0039] In the above formula, X rrepresents the real part of the LSTM input feature, X i Represents the imaginary part of the input feature of the long short-term memory network, LSTM r , LSTM i Respectively represent the long and short-term memory networks for processing the real and imaginary parts, F out Represents the output of the long short-term memory network.

[0040] Furthermore, in the complex spectrum convolutional recurrent neural network, a loss function is set to assist speech enhancement processing; the loss function is obtained by training the noise signal-to-noise ratio and the scale-invariant signal-to-noise ratio on the optimizer, and the loss function is expressed as:

[0041] L=L snr +λL sisnr ;

[0042]

[0043] In the above formula, L represents the loss function, L snr represents the noise signal-to-noise ratio, L sisnr represents the scale-invariant signal-to-noise ratio, λ represents the hyperparameter, S represents the real audio waveform, represents the target audio waveform output by the complex spectrum convolutional recurrent neural network, S m represents the mth segment of the real audio waveform in S, express The mth segment of target audio waveform.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention constructs auxiliary clue information by integrating multi-dimensional soil geological information and soil environmental information, including soil quality, soil water content, and surface vegetation conditions, and uses a clue mask network to comprehensively generate clue mask vectors. At the same time, in the feature extraction stage, a three-dimensional complex logarithmic Mel spectrum is selected, including the logarithmic Mel spectrum corresponding to the real and imaginary parts of the short-time Fourier transform, the first-order difference of the logarithmic Mel spectrum, and the second-order difference of the logarithmic Mel spectrum. The above feature information is input into a complex convolutional recursive neural network, and the speech signal noise reduction and enhancement processing is realized through training and testing of a large amount of distributed sensor sound wave data. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram showing the connection relationship between the spatiotemporal data acquisition system for distributed buried optical fibers and the distributed buried optical fiber acoustic sensing speech enhancement processing system of the present invention.

[0047] Figure 2 It is a flow chart of the present invention for extracting features from spatiotemporal sound wave data.

[0048] Figure 3 It is a flow chart of the present invention for constructing a clue mask network based on one-hot encoding.

[0049] Figure 4 It is a schematic diagram of the complex spectral convolutional recursive neural network of the present invention. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be clearly described below in conjunction with the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention, and all other embodiments obtained by ordinary technicians in the field without making creative work are within the protection scope of the present invention.

[0051] The present invention provides a distributed optical fiber sensing voice enhancement method based on soil geology and environmental information, comprising the following steps:

[0052] S1. Collecting space-time acoustic wave data using a space-time data acquisition system for distributed buried optical fiber. The space-time data acquisition system for distributed buried optical fiber is as follows: Figure 1 As shown, it includes a laser, an acousto-optic modulator, an erbium-doped fiber amplifier, a circulator, a distributed buried optical fiber, a phase demodulation unit, a photodetector and a space-time data acquisition card. The laser adopts a narrow-band laser. The laser emits a light source, which enters the acousto-optic modulator and is modulated into a pulse light by the acousto-optic modulator. The pulse light passes through the erbium-doped fiber amplifier, is subjected to method and enhancement, and then enters the circulator. After passing through the circulator, it is propagated through the distributed buried optical fiber. When a physical event occurs, that is, an acoustic wave signal is generated, the acoustic wave vibration will pass through the complex geological structure and finally act on the distributed buried optical fiber, causing the phase of the backward Rayleigh scattered light to change. The carried phase passes through the phase modulation unit and is then received by the photodetector. An analog-to-digital converter is set, and the analog-to-digital converter converts the phase received by the photodetector into a digital signal and sends it to the space-time data acquisition card, thereby collecting the space-time acoustic wave data.

[0053] S2: Enhance the spatiotemporal acoustic wave data collected in step S1, using Figure 1 The distributed buried optical fiber acoustic sensing speech enhancement processing system in the invention performs the enhancement operation of the spatiotemporal acoustic wave data; specifically comprises the following steps:

[0054] S21, such as Figure 2 As shown, feature extraction of the spatiotemporal acoustic wave data collected in step S1 is helpful to facilitate subsequent model training. Specifically, the following steps are included:

[0055] S211, acquiring the time domain signal of the space-time acoustic wave data from the space-time data acquisition card, and performing short-time Fourier transform to obtain the real part and imaginary part of the short-time Fourier transform; the specific steps are:

[0056] The noisy space-time sound wave data is represented as y, y=x+v, wherein x represents the denoised space-time sound wave data, and v represents the noise signal.

[0057] Perform short-time Fourier transform on both sides of the expression of noisy space-time sound wave data to obtain Y=X+V, where Y, X, and V are the short-time Fourier transforms of y, x, and v respectively.

[0058] Further expanded into the real and imaginary parts of the complex number, expressed as:

[0059] Y r +iY i =X r +iX i +V r +iV i ;

[0060] In the above formula, Y r represents the real part of Y, iY i represents the imaginary part of Y; X r represents the real part of X, iX i represents the imaginary part of X; V r represents the real part of V, iV i Represents the imaginary part of V.

[0061] S212, setting a Mel filter bank, passing all the real parts and imaginary parts obtained by the short-time Fourier transform in step S211 through the set Mel filter bank respectively, to obtain a Mel short-time Fourier transform spectrum.

[0062] S213, taking the logarithm of the Mel short-time Fourier transform spectrum obtained in step S212 to obtain a logarithmic Mel time spectrum diagram, which is specifically obtained by the following formula:

[0063]

[0064] In the above formula, M(f, t) represents the value of the Mel spectrogram at frequency f and time t, |H j (f)| represents the response value of the jth Mel filter at frequency f, |X j (t)| represents the amplitude of the frequency domain signal corresponding to the jth Mel filter at time t, n represents the total number of Mel filters, and i is 1-n.

[0065] S214. Perform first-order and second-order differences on the logarithmic Mel-spectrogram obtained in step S213 to obtain corresponding first-order differences of the logarithmic Mel spectrum and second-order differences of the logarithmic Mel spectrum, thereby obtaining the features to be extracted from the spatiotemporal sound wave data. The features to be extracted from the spatiotemporal sound wave data include the logarithmic Mel-spectrogram, the first-order differences of the logarithmic Mel spectrum, and the second-order differences of the logarithmic Mel spectrum.

[0066] S22, constructing a clue mask network based on soil geology and environmental information, specifically including the following steps:

[0067] S221. The soil geological information and the soil environmental information are integrated to obtain the corresponding one-hot coding. The soil geological information includes soil depth and soil water content. The soil environmental information includes surface vegetation conditions. The soil depth, soil water content and surface vegetation conditions can all be represented by numbers. The soil depth, soil water content and surface vegetation conditions represented by the data are converted into binary to form a one-hot coding. The soil depth, soil water content and surface vegetation conditions are represented by numbers as shown in Table 1:

[0068] Table 1

[0069]

[0070]

[0071] A first moisture content threshold and a second moisture content threshold are set, the first moisture content threshold is less than the second moisture content threshold, when the moisture content of the area is less than the first moisture content threshold, it is a low moisture content environment, when the moisture content of the area is greater than or equal to the first moisture content threshold and less than the second moisture content threshold, it is a medium moisture content environment, and when the moisture content of the area is greater than or equal to the second moisture content threshold, it is a high moisture content environment; a first unit vegetation threshold and a second unit vegetation threshold are set, the first unit vegetation threshold is less than the second unit vegetation threshold, when the number of unit vegetation in the area is less than the first unit vegetation threshold, it is no vegetation, when the number of unit vegetation in the area is greater than or equal to the first unit vegetation threshold and less than the second unit vegetation threshold, it is medium vegetation, and when the number of unit vegetation in the area is greater than the second unit vegetation threshold, it is dense vegetation.

[0072] S222, construct a clue mask network according to the unique hot encoding obtained in step S221; the specific process is as follows Figure 3 As shown, an encoder is set, and the encoder set is a Transformer encoder. The unique hot encoding is input into the encoder to obtain the output E of the encoder. k , in Represents the dimension of the encoder, and the output E of the encoder k Perform multiplication query integration with the one-hot encoding l to obtain E k ′, that is, E k ′=E k ⊙l, E k 、E k ′ are projected to the decoder dimension N using a one-dimensional convolutional layer d 、N e , respectively get the projection code P k , Pk ′, then, P k , P k ′ are respectively passed into the self-attention module and the cross-attention module, P k After being processed by the self-attention module, it passes through the feedforward neural network and is input into the cross-attention module to calculate the decoded representation and obtain the target mask in the projection decoder space. The target mask in the projection decoder space passes through the feedforward neural network and the one-dimensional convolution layer in turn, and is projected back to the encoder dimension to obtain m. k ′, Finally, the clue mask vector is obtained, thereby obtaining the clue mask network; the clue mask vector is calculated by the following formula:

[0073] m k =m k ′+E k ′;

[0074] In the above formula, m k Represents the clue mask vector.

[0075] S23, construct a complex spectrum convolutional recursive neural network, such as Figure 4 As shown, the logarithmic Mel-spectrogram, Mel-spectrogram first-order difference and logarithmic Mel-spectrogram second-order difference obtained in step S21 are input into the complex spectrum convolutional recursive neural network to achieve speech enhancement of spatiotemporal sound wave data.

[0076] The complex spectrum convolutional recurrent neural network includes a clue mask network and a backbone network. The backbone network includes an encoder, a long short-term memory network and a decoder, wherein the encoder and the decoder are connected through a U-Net for fast linking; the first-order difference of the Mel spectrum and the second-order difference of the logarithmic Mel spectrum are input into the encoder, the encoder processes and outputs them, the output of the encoder is input into the clue mask network to obtain a clue mask vector, and then the clue mask vector and the clue mask network are multiplied and integrated, and the result of the multiplication query integration is passed through the long short-term memory network and the decoder in turn, and the output of the decoder is the speech enhancement data of the spatiotemporal sound wave data.

[0077] The encoder consists of four complex two-dimensional convolutions, complex batch normalization operations, and real PReLU activation functions. The complex two-dimensional convolution uses a smooth transition hole convolution. The features after the complex two-dimensional convolution are then processed by complex batch normalization operations and real PReLU activation functions in turn. The specific expression of the complex two-dimensional convolution is:

[0078]

[0079] y[i] represents the output sequence, w[k] represents the convolution kernel, r represents the void ratio of the convolution kernel, K represents the size of the convolution kernel, k is 1-K, x[i-rk] represents the sequence after the input sequence is processed, and x[i] represents the input sequence.

[0080] The long short-term memory network is used to connect the encoder and the decoder. The input of the long short-term memory network is the product of the output of the encoder and the clue mask vector output by the clue mask network. The real part and the imaginary part of the input of the long short-term memory network are processed separately. The processing process is specifically expressed by the following formula:

[0081] F rr =LSTM r (X r );

[0082] F ir =LSTM r (X i );

[0083] F ri =LSTM i (X r );

[0084] F ii =LSTM i (X i );

[0085] F out =(F rr -F ii )+j(F ir +F ri );

[0086] In the above formula, X r represents the real part of the LSTM input feature, X i Represents the imaginary part of the input feature of the long short-term memory network, LSTM r , LSTM i Respectively represent the long and short-term memory networks for processing the real and imaginary parts, F out Represents the output of the long short-term memory network.

[0087] The decoder has the same structure as the encoder, and also includes 4 complex two-dimensional convolutions, complex batch normalization operations, and real PReLU activation functions. The main function of the decoder is to achieve feature dimensionality reduction, while adding the features previously in the encoder accordingly, and finally obtaining speech enhancement data of the spatiotemporal sound wave data.

[0088] In the complex spectrum convolutional recurrent neural network, a loss function is set to assist speech enhancement processing; the loss function is obtained by training the noise signal-to-noise ratio and the scale-invariant signal-to-noise ratio on the optimizer. The optimizer uses the Adam optimizer, and the loss function is expressed as:

[0089] L=L snr +λL sisnr ;

[0090]

[0091]

[0092] In the above formula, L represents the loss function, L snr represents the noise signal-to-noise ratio, L sisnr represents the scale-invariant signal-to-noise ratio, λ represents the hyperparameter, S represents the real audio waveform, represents the target audio waveform output by the complex spectrum convolutional recurrent neural network, S m represents the mth segment of the real audio waveform in S, express The mth segment of the target audio waveform.

[0093] The present invention constructs auxiliary clue information by integrating multi-dimensional soil geological information and soil environmental information, including soil quality, soil water content, and surface vegetation conditions, and uses a clue mask network to comprehensively generate clue mask vectors. At the same time, in the feature extraction stage, a three-dimensional complex logarithmic Mel spectrum is selected, including the logarithmic Mel spectrum corresponding to the real and imaginary parts of the short-time Fourier transform, the first-order difference of the logarithmic Mel spectrum, and the second-order difference of the logarithmic Mel spectrum. The above feature information is input into a complex convolutional recursive neural network, and the speech signal noise reduction and enhancement processing is realized through training and testing of a large amount of distributed sensor sound wave data.

[0094] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A distributed optical fiber sensing speech enhancement method based on soil geology and environmental information, characterized in that: The following steps are involved: S1. Collecting spatiotemporal acoustic wave data; S2, performing a speech enhancement operation on the spatiotemporal sound wave data collected in step S1, specifically comprising the following steps: S21, extracting features from spatiotemporal sound wave data; S22, constructing a clue mask network based on soil geological information and soil environmental information; S23. Construct a complex spectrum convolutional recurrent neural network, which includes a clue mask network and a backbone network. Input the features extracted from the spatiotemporal sound wave data in step S21 into the complex spectrum convolutional recurrent neural network to obtain speech enhancement data of the spatiotemporal sound wave data.

2. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 1 is characterized in that: S21 specifically includes the following steps: S211, performing short-time Fourier transform on the time domain signal of the space-time sound wave data to obtain the real part and the imaginary part of the short-time Fourier transform; S212, setting a Mel filter bank, passing the real part and the imaginary part obtained in step S211 through the Mel filter bank respectively, to obtain a Mel short-time Fourier transform spectrum; S213, taking the logarithm of the Mel short-time Fourier transform spectrum to obtain a logarithmic Mel time spectrum diagram; S214. Perform first-order differences and second-order differences on the logarithmic Mel-spectrogram to obtain corresponding first-order differences of the logarithmic Mel spectrum and second-order differences of the logarithmic Mel spectrum, thereby obtaining features to be extracted from the spatiotemporal sound wave data. The features to be extracted from the spatiotemporal sound wave data include the logarithmic Mel-spectrogram, the first-order differences of the logarithmic Mel spectrum, and the second-order differences of the logarithmic Mel spectrum.

3. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 2 is characterized in that: In step S213, the logarithmic Mel-time spectrum is obtained by the following formula: In the above formula, M(f,t) represents the value of the Mel spectrogram at frequency f and time t, |H j (f)| represents the response value of the jth Mel filter at frequency f, |X j (t)| represents the amplitude of the frequency domain signal corresponding to the jth Mel filter at time t, n represents the total number of Mel filters, and j is 1-n.

4. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 1 is characterized in that: S22 specifically includes the following steps: S221, performing feature fusion of soil geological information and soil environmental information, the soil geological information includes soil depth and soil water content, the soil environmental information includes surface vegetation conditions, performing binary conversion on soil depth, soil water content, and surface vegetation conditions to obtain a one-hot encoding; S222. Construct a clue mask network based on the one-hot encoding.

5. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 4 is characterized in that: The specific method of constructing the clue mask network in step S222 is: setting a Transformer encoder, setting the one-hot encoding representation as l; inputting l into the Transformer encoder to obtain the output E of the Transformer encoder k , E k Integrate with l to perform multiplication query and get E k ′, E k 、E k ′ are projected to the decoder dimension N using a one-dimensional convolutional layer d 、N e , respectively get the projection code P k , P k ′, then, P k , P k ′ are respectively passed into the self-attention module and the cross-attention module, P k After being processed by the self-attention module and then passed through the feedforward neural network, it is input into the cross-attention module. The output of the cross-attention module is the target mask in the projection decoder space. The target mask passes through the feedforward neural network and the one-dimensional convolution layer in turn, and is projected back to the Transformer encoder dimension to obtain m k ′, and finally the clue mask vector is obtained, thus obtaining the clue mask network.

6. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 5 is characterized in that: The clue mask vector is calculated as follows: m k =m k ′+E k ′; In the above formula, m k Represents the clue mask vector.

7. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 5 or 6 is characterized in that: The backbone network includes an encoder, a long short-term memory network and a decoder. The encoder and the decoder are quickly linked through the long short-term memory network. In step S21, the features extracted from the spatiotemporal sound wave data are input into the encoder, and the encoder processes and outputs the features. The output of the encoder is input into the clue mask network to obtain a clue mask vector, and then the clue mask vector and the clue mask network are multiplied and integrated. The result of the multiplication query integration passes through the long short-term memory network and the decoder in turn, and the output of the decoder is the speech enhancement data of the spatiotemporal sound wave data.

8. The distributed optical fiber sensing voice enhancement method based on soil geology and environmental information according to claim 7 is characterized in that: The encoder and the decoder both include 4 complex two-dimensional convolutions, complex batch normalization operations, and real PReLU activation functions. The specific expression of the complex two-dimensional convolution is: y[i] represents the output sequence, w[k] represents the convolution kernel, r represents the void ratio of the convolution kernel, K represents the size of the convolution kernel, k is 1-K, x[i-rk] represents the sequence after the input sequence is processed, and x[i] represents the input sequence.

9. The distributed optical fiber sensing speech enhancement method based on soil geology and environmental information according to claim 7 is characterized in that: The long short-term memory network processes the real and imaginary parts of the input separately. The processing process is specifically expressed by the following formula: F rr =LSTM r (X r ); F ir =LSTM r (X i ); F ri =LSTM i (X r ); F ii =LSTM i (X i ); F out =(F rr -F ii )+j(F ir +F ri ); In the above formula, X r represents the real part of the LSTM input feature, X i Represents the imaginary part of the input feature of the long short-term memory network, LSTM r , LSTM i Respectively represent the long and short-term memory networks for processing the real and imaginary parts, F out Represents the output of the long short-term memory network.

10. The distributed optical fiber sensing voice enhancement method based on soil geology and environmental information according to claim 7 is characterized in that: In the complex spectrum convolutional recurrent neural network, a loss function is set to assist speech enhancement processing; the loss function is obtained by training the noise signal-to-noise ratio and the scale-invariant signal-to-noise ratio on the optimizer, and the loss function is expressed as: L=L snr +λL sisnr ; In the above formula, L represents the loss function, L snr represents the noise signal-to-noise ratio, L sisnr represents the scale-invariant signal-to-noise ratio, λ represents the hyperparameter, S represents the real audio waveform, represents the target audio waveform output by the complex spectrum convolutional recurrent neural network, S m represents the mth segment of the real audio waveform in S, express The mth segment of the target audio waveform.

Images