A Digital Audio Processing Method and System for Audio Equipment

Abstract

The present invention relates to the technical field of digital audio processing, and particularly to a digital audio processing method and system for a sound system. The system includes an audio signal acquisition module, a signal analysis module, an adaptive adjustment module, a howling suppression module, and a sound effect optimization feedback module. The audio signal acquisition module collects data such as the spectrum and amplitude of the sound system in real time. The signal analysis module generates a signal analysis report based on the non-negative matrix factorization algorithm and the generative adversarial network algorithm. The adaptive adjustment module automatically adjusts audio parameters through the fuzzy logic control algorithm. The howling suppression module uses the adaptive feedback cancellation algorithm to suppress howling noise. The sound effect optimization feedback module enhances the sound effect using a Bezier curve to achieve frequency, phase, and spatial positioning optimization. Through the collaborative processing of audio acquisition, precise analysis, adaptive adjustment, howling suppression, and spatial sound effect optimization, the present invention not only improves the quality and stability of the audio, but also enhances the spatial effect and user experience of the audio.

Description

Technical Field

[0001] The present invention relates to the technical field of digital audio processing, and particularly to a digital audio processing method and system for audio equipment. Background Art

[0002] In modern audio devices, the quality of sound directly affects the user's auditory experience. With the rapid development of digital technology, digital audio processing has gradually become an important means to improve the sound quality in audio equipment. Traditional audio systems mainly rely on analog circuits to process audio signals, which have certain limitations in terms of noise, distortion, and frequency response, and are difficult to meet the user's demand for high-fidelity sound quality. In recent years, digital audio processing systems have enhanced the audio output effect by applying digital signal processing technologies (such as filtering, equalization, echo cancellation, reverberation, etc.). There are still the following problems: existing systems mostly rely on fixed algorithms or manual adjustment, and cannot automatically adapt to different environments to achieve the best sound effect; traditional suppression algorithms cannot eliminate howling noise in real-time dynamically, affecting the user experience; existing systems have limited performance in spatial positioning and sound effect enhancement, and it is difficult to achieve an immersive experience. Summary of the Invention

[0003] To solve the above problems, the present invention provides a digital audio processing method and system for audio equipment, which solves the problem of how to design a digital audio processing system in an audio system that can automatically adapt to different environments to adjust audio parameters, suppress howling noise in real-time dynamically, and at the same time improve spatial positioning and sound effect enhancement performance, thereby not only improving the quality and stability of audio, but also enhancing the spatial effect of audio and the user experience.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] On the one hand, a digital audio processing system for audio equipment includes an audio signal acquisition module, a signal analysis module, an adaptive adjustment module, a howling suppression module, and a sound effect optimization feedback module that are communicatively connected in sequence;

[0006] The audio signal acquisition module is used to collect audio signal data of the audio equipment in real-time through a sensor network, and the audio signal data includes spectrum, amplitude, phase, sound pressure level, and reverberation time;

[0007] The signal analysis module is used to perform spectral feature decomposition on the audio signal data based on the non-negative matrix factorization algorithm, analyze the spectral features using the generative adversarial network algorithm, compare with the standard spectral curve, and generate a signal analysis report;

[0008] The adaptive adjustment module is used to automatically adjust the gain, frequency response, and equalization parameters of the audio signal based on the signal analysis report using the fuzzy logic control algorithm;

[0009] The howling suppression module is used to detect the howling noise components in the audio signal in real time by using the adaptive feedback cancellation algorithm and perform dynamic suppression;

[0010] The sound effect optimization feedback module is used to optimize the frequency, phase and spatial positioning of the audio signal by constructing a high-order Bessel curve model of multi-channel audio based on the spatial sound effect enhancement algorithm fitted by the high-order Bessel curve.

[0011] Furthermore, the operation process of the signal analysis module includes the following steps:

[0012] Receive the audio signal data and perform preprocessing operations;

[0013] Use the non-negative matrix factorization algorithm to decompose the spectrum features of the preprocessed audio signal data, generate multiple independent spectrum bases and corresponding weight matrices, extract the spectrum feature information of different frequency bands, and mark the main frequency components of the signal;

[0014] Adopt the generative adversarial network algorithm to perform multi-layer comparison between the spectrum features and the built-in standard spectrum curve of the system, generate a high-precision feature matching model through adversarial training, identify and highlight abnormal spectrum components, and generate the stability and similarity scores of the spectrum features at the same time;

[0015] Import the analysis results into the dynamic adjustment model, output the feature difference map based on the spectrum feature report generated by the generative adversarial network algorithm, and generate a dynamic adjustment prompt according to the spectrum difference and stability score;

[0016] Integrate the comparison results and dynamic adjustment detection information to generate a signal analysis report, including the detailed spectrum component change trend, abnormal frequency identification results, gain distribution suggestions, frequency response curve matching degree, and the recommended adjustment range for each audio parameter.

[0017] Even further, the construction process of the non-negative matrix factorization algorithm includes the following steps:

[0018] Divide the audio signal data into several adaptive sub-bands through the intelligent frequency band division algorithm, extract the main spectrum features according to the characteristics of different frequency bands, and form an initial feature matrix;

[0019] Initialize the basis matrix by using the sparsity control mechanism, and adjust the structure of the basis matrix through a unique sparse optimization function;

[0020] In the decomposition iteration, dynamically adjust the sparse constraint parameters and residual suppression weights to control the noise influence in the decomposition process, and gradually optimize the sparse structure of the spectrum basis matrix;

[0021] Integrate the final basis matrix and the weight matrix for output, form a complete set of spectral features through multi-band feature summarization, and generate differential feedback of the spectral features.

[0022] Further, the formula of the fuzzy logic control algorithm is as follows:

[0023] ;

[0024] Wherein, represents the optimal gain adjustment amount; and respectively represent the weights of the spectral features and the equalization features in the overall adjustment; n represents the number of fuzzy rules, corresponding to different frequency band characteristics; represents the weight of the i-th fuzzy rule; represents the spectral features calculated according to the input variable x; represents the spectral feature weight function; and represent the integration interval; m represents the number of equalization features; represents based on the input variable calculated equalization features.

[0025] Further, the operation process of the howling suppression module includes the following steps:

[0026] Detect the howling components in the audio signal in real time through the adaptive feedback cancellation algorithm, identify and locate the main frequencies that cause howling, and simultaneously construct a feedback path model;

[0027] Utilize the feedback path model, apply an adaptive filter to generate a suppression signal that is out of phase with the howling frequency, and dynamically eliminate the howling noise components by adjusting the filter weight coefficients in real time;

[0028] According to the change of the howling frequency, adaptively adjust the gain and frequency band range of the filter, and continuously optimize the feedback path model through the adaptive feedback cancellation algorithm to enhance the suppression effect;

[0029] After completing the feedback suppression, generate a howling suppression report, including the change trend of the howling frequency, the feedback path stability score, the suppression effect of each frequency band, and optimization suggestions for the feedback path model.

[0030] Even further, the formula of the feedback path model is as follows:

[0031] ;

[0032] Wherein, represents the output audio signal after suppressing howling; represents the input original audio signal; N represents the number of feedback paths; Denotes the adaptive gain control coefficient of the $I$-th feedback path; Denotes the influence degree of different paths on the feedback signal; Denotes the delay time of the $i$-th feedback path; Denotes the suppression intensity coefficient; Denotes the amplitude power exponent of the feedback signal; Denotes the amplitude of the feedback signal in the $i$-th feedback path.

[0033] Furthermore, the operation process of the sound effect optimization feedback module includes the following steps:

[0034] Collect the spectrum, phase and spatial positioning data of the audio through a multi-channel sensor, use the high-order Bessel curve fitting algorithm to generate a three-dimensional spatial sound effect model, and convert the spatial distribution characteristics of the sound source into a multi-order control point matrix;

[0035] Based on the control point matrix generated by modeling, fine-tune the frequency response and phase of each frequency band;

[0036] Utilize the multi-channel feedback signal to adaptively regulate the Bessel curve control parameters and dynamically adjust the spatial distribution of the audio signal;

[0037] Generate an optimization feedback report according to the result of the sound effect optimization, including the gain distribution map of the spatial sound effect enhancement, the error analysis of each sound source positioning, the optimized frequency response curve and the recommended parameter adjustment range.

[0038] Even further, the formula of the three-dimensional spatial sound effect model is as follows:

[0039] ;

[0040] Wherein, Denotes the positioning point of the audio signal in the three-dimensional space at time $t$, frequency $f$ and phase ; Denotes the $i$-th basis function of the Bessel curve; Denotes the $i$-th control point coordinate matrix of the Bessel curve; Denotes the frequency parameter of the $i$-th frequency band; Denotes the phase offset of the $i$-th frequency band; Denotes the attenuation coefficient, controlling the influence of the distance on the audio signal; Denotes the dynamic distance function of the control point at time $t$; $t$ denotes the time parameter; $n$ denotes the order of the Bessel curve.

[0041] On the other hand, a digital audio processing method for an audio device includes the following steps:

[0042] Collect the audio signal data of the sound system in real time through the sensor network;

[0043] Perform spectral feature decomposition on the audio signal data based on the non - negative matrix factorization algorithm, and use the generative adversarial network algorithm to analyze the spectral features, compare with the standard spectral curve, and generate a signal analysis report;

[0044] Based on the signal analysis report, use the fuzzy logic control algorithm to automatically adjust the gain, frequency response, and equalization parameters of the audio signal;

[0045] Use the adaptive feedback cancellation algorithm to detect the howling noise component in the audio signal in real time and perform dynamic suppression;

[0046] Based on the spatial sound effect enhancement algorithm of high - order Bessel curve fitting, by constructing a high - order Bessel curve model for multi - channel audio, optimize the frequency, phase, and spatial positioning of the audio signal.

[0047] The beneficial effects of the present invention are as follows: The audio signal acquisition module of the present invention can collect multi - dimensional audio signal data such as spectrum, amplitude, and phase in real time through the sensor network, ensuring the comprehensiveness and accuracy of the data, and providing high - quality input for subsequent processing. The signal analysis module realizes spectral feature decomposition through the non - negative matrix factorization algorithm, and combines the generative adversarial network to deeply analyze the spectral features. This combined analysis method can more accurately identify the characteristics of the audio signal, realize intelligent comparison with the standard spectrum, quickly generate a signal analysis report, and improve the accuracy and efficiency of the analysis. The adaptive adjustment module uses the fuzzy logic control algorithm to automatically adjust the gain, frequency response, and equalization parameters of the audio. Through the intelligent adjustment mechanism, the system can adapt to different environments, optimize the output effect of the audio, and make the sound effect more consistent and of higher quality. The howling suppression module uses the adaptive feedback cancellation algorithm, which can quickly perform dynamic suppression when detecting howling noise in real time, effectively reduce the howling interference in the audio, and improve the user's auditory experience. The sound effect optimization feedback module uses the spatial sound effect enhancement algorithm based on high - order Bessel curve fitting. The system constructs a spatial positioning model for multi - channel audio, making the frequency, phase, and spatial positioning of the audio more accurate and three - dimensional, and significantly enhancing the spatial sense and immersive experience of the sound system. Brief Description of the Drawings

[0048] Figure 1 is a schematic diagram of the modules of a digital audio processing system for a sound system according to the present invention.

[0049] Figure 2 is a schematic flowchart of the operation process of the howling suppression module provided by an embodiment of the present invention.

[0050] Figure 3 is a schematic flowchart of a digital audio processing method for a sound system according to the present invention. Detailed implementation mode

[0051] Please refer to Figures 1-3 As shown, the present invention relates to a digital audio processing method and system for audio equipment.

[0052] Embodiment 1

[0053] A digital audio processing system for audio equipment, comprising an audio signal acquisition module, a signal analysis module, an adaptive adjustment module, a howling suppression module, and a sound effect optimization feedback module that are communicatively connected in sequence;

[0054] The audio signal acquisition module is used to collect audio signal data of the audio equipment in real time through a sensor network, and the audio signal data includes spectrum, amplitude, phase, sound pressure level, and reverberation time;

[0055] It should be noted that the real-time collection of audio signal data of the audio equipment through the sensor network is as follows:

[0056] Sensor type: A variety of sensors are arranged in the audio system, including but not limited to microphones, pressure sensors, accelerometers, etc., to ensure that the spectrum, amplitude, and other signal characteristics of sound are captured from different angles.

[0057] Microphone array design: By configuring microphones with different numbers and positions to form a microphone array, the system can collect audio information from different directions, distinguish direct sound and reflected sound, and obtain high-quality environmental sound data.

[0058] Sound source position optimization: According to the usage environment of the audio equipment, optimize the installation position of the sensors. For example, in a multi-channel system, the sensors can be placed in the corners of multiple rooms to monitor the sound field conditions in different areas. This arrangement can improve the coverage and accuracy of audio collection.

[0059] Before the collection starts, calibrate the noise of the sensor collection environment to establish a background noise baseline. By real-time monitoring and comparison, identify the changes in environmental noise, thereby improving the stability of signal collection. Regularly self-check and calibrate parameters such as the sensitivity and frequency response of the sensors to maintain the accuracy and stability of the sensors. For example, the frequency response of the sensors can be calibrated through an automated white noise test or a signal generator. Through the feedback mechanism of the collection module, detect the quality and signal strength of the collected data in real time. When the signal deviates from the preset range, the system will automatically adjust the collection parameters or issue a warning to ensure the accuracy and reliability of data collection.

[0060] The signal analysis module is used to perform spectral feature decomposition on the audio signal data based on the non - negative matrix factorization algorithm, and analyze the spectral features by using the generative adversarial network algorithm, compare with the standard spectral curve, and generate a signal analysis report;

[0061] Among them, the operation process of the signal analysis module includes the following steps:

[0062] Receive the audio signal data and perform pre - processing operations;

[0063] Use the non - negative matrix factorization algorithm to perform spectral feature decomposition on the pre - processed audio signal data, generate multiple independent spectral bases and corresponding weight matrices, extract spectral feature information in different frequency bands, and mark the main frequency components of the signal;

[0064] Adopt the generative adversarial network algorithm, conduct multi - layer comparison of the spectral features with the standard spectral curve built in the system, generate a high - precision feature matching model through adversarial training, identify and highlight abnormal spectral components, and simultaneously generate stability and similarity scores of the spectral features;

[0065] Specifically, adopt the generative adversarial network (GAN) algorithm. The generator is used to simulate the standard spectral features, and the discriminator is used to compare the input spectral features with the generated standard spectrum. The spectral features obtained by NMF decomposition are compared with the standard spectral curve built in the system at multiple levels. Through the adversarial training of the generator and the discriminator, the GAN model gradually optimizes its feature recognition accuracy. The closer the features generated by the generator are to the standard spectral curve, the easier it is for the discriminator to identify abnormal frequency components. Through multi - layer comparison, GAN can identify the abnormal spectral components in the audio signal and highlight them. The GAN model generates a similarity score according to the feature matching degree, and the stability score reflects the smoothness of the audio signal features over time. The scoring results are used to quantify the stability of the current signal features and the matching degree with the standard curve.

[0066] Import the analysis results into the dynamic adjustment model, output a feature difference map based on the spectral feature report generated by the generative adversarial network algorithm, and generate a dynamic adjustment prompt according to the spectral difference and stability score;

[0067] Specifically, input the spectral feature analysis results obtained by GAN comparison into the dynamic adjustment model for further processing. The dynamic adjustment model generates a feature difference map, visually displays the difference between the current spectral features and the standard spectrum, and helps to identify the frequency bands that need to be adjusted. According to the spectral difference, stability score and similarity score, generate dynamic adjustment suggestions such as gain and equalization for specific frequency bands. For example, if the high - frequency part of the spectrum deviates from the standard, the system will recommend reducing the high - frequency gain to ensure the sound effect output quality.

[0068] Generate a signal analysis report by comprehensively comparing the results with the dynamically adjusted detection information, including the detailed trend of spectral component changes, the abnormal frequency identification results, the gain distribution suggestions, the frequency response curve matching degree, and the recommended adjustment ranges for each audio parameter.

[0069] The construction process of the non - negative matrix factorization algorithm includes the following steps:

[0070] Divide the audio signal data into several adaptive sub - bands through an intelligent frequency - band division algorithm, extract the main spectral features according to the characteristics of different frequency bands, and form an initial feature matrix;

[0071] Specifically, according to the frequency range of the audio signal and application requirements, use an intelligent frequency - band division algorithm to divide the audio signal into multiple adaptive sub - bands. For example, in the speech processing scenario, divide it into low - frequency (below 250 Hz), mid - frequency (250 - 4000 Hz), and high - frequency (above 4000 Hz) sub - bands to adapt to the auditory sensitivity of the human ear. Extract the main spectral features in each adaptive sub - band to form an initial feature matrix. This matrix is used to describe the energy distribution of the audio signal in each frequency band.

[0072] Initialize the basis matrix using a sparsity control mechanism and adjust the structure of the basis matrix through a unique sparse optimization function;

[0073] Specifically, initialize the basis matrix through a sparsity control mechanism. The sparsity design of the basis matrix helps to extract the significant spectral features in the audio signal and reduce the interference of non - significant components. Design a dedicated sparse optimization function to control the sparsity of the structure of the basis matrix, so that the basis matrix can centrally reflect the main frequency components in the signal.

[0074] During the decomposition iteration, dynamically adjust the sparse constraint parameter and the residual suppression weight to control the noise impact during the decomposition process, and gradually optimize the sparse structure of the spectral basis matrix;

[0075] Specifically, during the iteration of non - negative matrix factorization, dynamically adjust the sparse constraint parameter to ensure the optimization of the sparsity of the basis matrix at different iteration stages. According to the error in the decomposition result, increase the residual suppression weight to control the noise impact. For example, when the error is large, increase the residual suppression weight to reduce the influence of noise on the basis matrix. Through the dynamic adjustment of sparse constraint and residual suppression, gradually optimize the sparse structure of the basis matrix, so that the finally decomposed basis matrix is more representative and reflects the main features of the audio signal.

[0076] Integrate and output the final basis matrix and the weight matrix, form a complete spectral feature set through multi - band feature summary, and generate a differential feedback of the spectral features.

[0077] The adaptive adjustment module is used to automatically adjust the gain, frequency response, and equalization parameters of the audio signal based on the signal analysis report by using a fuzzy logic control algorithm;

[0078] Specifically, according to the signal analysis report, the main parameters affecting the audio output are selected as the inputs of the fuzzy logic controller. These input parameters include spectral deviation, signal amplitude, background noise level, etc.

[0079] Spectral deviation: Describes the difference between the current audio signal and the standard spectrum, and is used to identify uneven frequency bands.

[0080] Signal amplitude: Used to judge whether the loudness of the audio signal is appropriate, and to avoid the volume being too large or too small.

[0081] Background noise level: By detecting the intensity of the background noise, the controller can adjust the gain parameter to improve the sound quality in a noisy environment.

[0082] Fuzzy rule base: Based on the input parameters, a series of fuzzy rules are designed. The fuzzy rule base includes some basic adjustment criteria, such as:

[0083] If the spectral deviation is large and the high-frequency band is strong, then reduce the high-frequency gain.

[0084] If the signal amplitude is small and the low-frequency band is weak, then increase the low-frequency gain.

[0085] If the background noise level is high and the signal amplitude is low, then moderately increase the gain to ensure the clarity of the output.

[0086] Fuzzy inference mechanism: Using the fuzzy inference method, the output decision is generated according to the input fuzzy values and the fuzzy rule base.

[0087] Furthermore, the formula of the fuzzy logic control algorithm is as follows:

[0088]

[0089] Wherein, represents the optimal gain adjustment amount, which is used to automatically adjust the gain of the audio signal to ensure that the audio effect reaches the best balance; and respectively represent the weights of the spectral characteristics and the equalization characteristics in the overall adjustment; n represents the number of fuzzy rules, corresponding to different frequency band characteristics; represents the weight of the i-th fuzzy rule; represents the spectral characteristics calculated according to the input variable x; represents the spectral characteristic weight function; and represent the integration interval, the frequency range to which the i-th fuzzy rule applies; m represents the number of equalization characteristics; Indicates the balanced features calculated based on the input variables

[0090] The howling suppression module is used to detect the howling noise components in the audio signal in real time by using an adaptive feedback cancellation algorithm and perform dynamic suppression;

[0091] Among them, the operation process of the howling suppression module includes the following steps:

[0092] Detect the howling components in the audio signal in real time through the adaptive feedback cancellation algorithm, identify and locate the main frequencies causing howling, and at the same time construct a feedback path model;

[0093] Specifically, the audio signal is monitored in real time through the adaptive feedback cancellation algorithm to identify the howling noise components. Use a spectrum analysis tool (such as fast Fourier transform) to decompose the audio signal and identify the frequencies with significant peaks and stability. These peaks are usually the main components of howling. For the detected spectral peaks, further analyze their frequency and amplitude characteristics to locate the main frequencies generating howling. Howling usually occurs in the higher frequency range, and the amplitude of this frequency component is relatively prominent. According to the howling frequency and its spectral characteristics, use the adaptive feedback cancellation algorithm to construct a feedback path model. The feedback path model is used to describe the acoustic feedback path between the microphone, speaker and the environment, and record the main frequencies, gains and delays and other characteristics in this path, providing a basis for subsequent howling suppression.

[0094] Utilize the feedback path model, apply an adaptive filter to generate a suppression signal that is out of phase with the howling frequency, and dynamically eliminate the howling noise components by adjusting the filter weight coefficients in real time;

[0095] Specifically, set an adaptive filter to generate a suppression signal that is out of phase with the detected howling frequency. The initial weight coefficients of the filter are determined by the parameters of the feedback path model, and the center frequency of the filter is set to the howling frequency in the initial stage. The suppression signal generated by the adaptive filter has the same frequency but opposite phase to the howling frequency, and is used to cancel the amplitude of the howling noise. This out-of-phase suppression signal is consistent with the howling noise in frequency but has the opposite amplitude, thus achieving an effective cancellation effect. The adaptive filter adjusts its weight coefficients in real time according to the audio signal to ensure that the frequency and amplitude of the out-of-phase signal always match the howling frequency. This process can use adaptive algorithms such as the least mean square error (LMS) algorithm or the RLS algorithm to ensure that the filter can also quickly adjust its output signal when the howling frequency changes.

[0096] According to the change situation of the howling frequency, adaptively adjust the gain and frequency band range of the filter, and continuously optimize the feedback path model through the adaptive feedback cancellation algorithm to enhance the suppression effect;

[0097] ​It should be noted that during the howling suppression process, the audio signal is monitored in real time to identify the changes in the howling frequency. When frequency drift is detected, the center frequency of the filter is adjusted to keep it consistent with the new howling frequency. According to the intensity and spectral peak of the howling signal, the gain parameter of the filter is dynamically adjusted to ensure the stability of the suppression effect. When the howling intensity is large, the filter gain is increased to more strongly suppress the howling; when the howling signal weakens or disappears, the gain is appropriately reduced to minimize the impact on the normal audio signal. The feedback path model is continuously adjusted to optimize the frequency band range of the filter. The frequency band range is finely tuned through an adaptive algorithm to avoid excessive suppression of too many frequency bands by the filter, thus ensuring the minimum interference to the normal audio signal. Based on the changes in the howling frequency and the adjustment process of the adaptive filter, the feedback path model is continuously optimized, and the latest howling frequency, gain, delay and other information are recorded. This dynamic feedback mechanism helps the system maintain an efficient howling suppression effect in a changing audio environment.

[0098] After the feedback suppression is completed, a howling suppression report is generated, including the change trend of the howling frequency, the stability score of the feedback path, the suppression effect of each frequency band, and optimization suggestions for the feedback path model.

[0099] Furthermore, the formula of the feedback path model is as follows:

[0100] ;

[0101] Where, represents the output audio signal after howling suppression; represents the original input audio signal; N represents the number of feedback paths; represents the adaptive gain control coefficient of the i-th feedback path, which controls the gain of howling suppression; represents the influence degree of different paths on the feedback signal; represents the delay time of the i-th feedback path, that is, the delay of the signal on this path. The delay time is calculated by the signal analysis module according to the reverberation time of the acoustic environment to adapt to different spatial sound effect characteristics; represents the suppression intensity coefficient, which controls the intensity of feedback suppression; represents the amplitude power exponent of the feedback signal, which is used for non-linear adjustment of the feedback signal to make the high-intensity howling feedback get a stronger suppression effect. Usually, the value is 2 or 3 to enhance the dynamic response of feedback cancellation; represents the amplitude of the feedback signal in the i-th feedback path.

[0102] Where, The specific formula is as follows:

[0103] Where, represents the maximum gain control coefficient; represents the feedback sensitivity coefficient; represents the amplitude of the feedback signal of the i-th feedback path, indicating the howling component detected in this path.

[0104] The sound effect optimization feedback module is used for a spatial sound effect enhancement algorithm based on high-order Bessel curve fitting. By constructing a high-order Bessel curve model for multi-channel audio, it optimizes the frequency, phase, and spatial positioning of the audio signal;

[0105] Among them, the operation process of the sound effect optimization feedback module includes the following steps:

[0106] Collect the spectrum, phase, and spatial positioning data of the audio through multi-channel sensors, use the high-order Bessel curve fitting algorithm to generate a three-dimensional spatial sound effect model, and convert the spatial distribution characteristics of the sound source into a multi-order control point matrix;

[0107] It should be noted that using the phase difference and time difference collected by the multi-channel sensors, calculate the spatial positioning of each sound source. Precise the spatial position of the sound source through the triangulation algorithm to obtain the coordinate information of each sound source in the three-dimensional space. According to the data collected by the multi-channel, use the high-order Bessel curve fitting algorithm to generate a three-dimensional spatial sound effect model. In specific implementation, describe the spatial distribution characteristics of the sound source through the multi-order control points of the Bessel curve, and convert the spatial positioning of the sound source into a multi-order control point matrix. According to the frequency, phase, and position characteristics of the sound source, generate the control point matrix of the Bessel curve. The control point matrix describes the distribution and change trend of the sound source in space, providing a basis for subsequent frequency response and phase optimization.

[0108] Based on the control point matrix generated by the modeling, fine-tune the frequency response and phase of each frequency band;

[0109] Specifically, based on the control point matrix, fine-tune the frequency response of each frequency band. By adjusting the position and weight of the control points, change the frequency response of the signals in different frequency bands to make it more in line with the goal of spatial sound effect enhancement. For example, increase the density of control points in the low-frequency band to enhance the immersion of the low-frequency; increase the smoothness of control points in the high-frequency band to improve the clarity. According to the control point matrix, gradually adjust the phase of each frequency band to make the spatial positioning of the sound source more accurate. The phase adjustment can be achieved by changing the phase distribution of the control points in the Bessel curve to achieve the optimization effect of the spatial sound effect. For example, in the overlapping area of multiple sound sources, separate the sound sources by adjusting the phase to avoid the deterioration of the sound quality caused by phase conflict. During the fine-tuning process, maintain the overall consistency of the spatial sound effect to ensure that the audio signals in different frequency bands maintain a natural transition in frequency response and phase. By smoothing the control point distribution, ensure that the sound effect model will not produce abrupt sound quality changes due to fine-tuning.

[0110] Using multi-channel feedback signals, adaptively regulate the control parameters of the Bessel curve, and dynamically adjust the spatial distribution of the audio signal;

[0111] Generate an optimized feedback report based on the results of the sound effect optimization, including the gain distribution map of the spatial sound effect enhancement, the error analysis of the positioning of each sound source, the optimized frequency response curve, and the recommended parameter adjustment range.

[0112] Furthermore, the formula of the three-dimensional spatial sound effect model is as follows:

[0113]

[0114] Where, represents the positioning point of the audio signal in the three-dimensional space at time t, frequency f, and phase , which is used to describe the distribution of the sound effect at different spatial positions and form the effect of three-dimensional sound effect enhancement; represents the i-th basis function of the Bessel curve, which is used to control the weights of different control points ; represents the coordinate matrix of the i-th control point of the Bessel curve, which is used to define the three-dimensional position of the spatial sound effect enhancement and ensure that the spatial distribution characteristics of the audio signal match the control points; represents the frequency parameter of the i-th frequency band; represents the phase offset of the i-th frequency band; represents the attenuation coefficient, which controls the influence of the distance on the audio signal and is used to simulate the spatial attenuation effect of the sound source in the three-dimensional space model; represents the dynamic distance function of the control point at time t, that is, the distance between the control point and the listening position, which is deduced through the sound pressure level and reverberation time parameters; t represents the time parameter, with a range in [0, 1], which controls the curve generation process and is synchronized with the time characteristics of the audio signal; n represents the order of the Bessel curve.

[0115] Embodiment 2

[0116] A digital audio processing method for an audio device, comprising the following steps:

[0117] Collect the audio signal data of the audio device in real time through a sensor network;

[0118] Perform spectral feature decomposition on the audio signal data based on the non-negative matrix factorization algorithm, and analyze the spectral features using the generative adversarial network algorithm, compare with the standard spectral curve, and generate a signal analysis report;

[0119] Based on the signal analysis report, the gain, frequency response, and equalization parameters of the audio signal are automatically adjusted using a fuzzy logic control algorithm;

[0120] An adaptive feedback cancellation algorithm is used to detect the howling noise component in the audio signal in real time and perform dynamic suppression;

[0121] Based on a spatial sound effect enhancement algorithm based on high-order Bessel curve fitting, by constructing a high-order Bessel curve model for multi-channel audio, the frequency, phase, and spatial positioning of the audio signal are optimized.

[0122] In this embodiment, a digital audio processing method for a sound system is applied to the digital audio processing system described in Embodiment 1, which will not be elaborated here.

[0123] In summary, through the multi-sensor array and optimized layout, the audio signal acquisition module can collect high-quality audio data in real time from multiple directions and positions, covering different sound field areas, improving the comprehensiveness and accuracy of the data. The non-negative matrix factorization algorithm is used for spectral feature decomposition, and the features are compared at multiple levels through a generative adversarial network to accurately identify and highlight abnormal spectral components. The generated spectral feature report provides a reliable basis for further dynamic adjustment of the audio signal.

[0124] The adaptive adjustment module of the present invention uses a fuzzy logic control algorithm to dynamically adjust the gain, frequency response, and equalization parameters of the audio signal based on the analysis report, and automatically generates an optimized adjustment plan to ensure the balance, clarity, and comfort of the sound effect output. The howling suppression module uses an adaptive feedback cancellation algorithm to detect and suppress the howling noise component in real time, dynamically adjusts the filter parameters, eliminates the howling component in the audio, and effectively enhances the stability and sound quality of the system. Through the high-order Bessel curve fitting algorithm, the sound effect optimization feedback module constructs a three-dimensional spatial sound effect model, dynamically optimizes the frequency response and phase of the audio signal, and realizes the accurate positioning and enhancement of the sound source in the three-dimensional space. The optimized spatial sound effect experience provides a more immersive auditory effect for users.

[0125] The system automatically calibrates the sensors and audio signals, detects the signal quality and provides a feedback mechanism, and adjusts the acquisition and processing parameters in real time. Through the analysis and optimization report, the system can adaptively control the key parameters to make the audio processing process more stable and intelligent.

[0126] The above embodiments are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary engineering and technical personnel in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A digital audio processing system for a sound system, characterized in that, It includes an audio signal acquisition module, a signal analysis module, an adaptive adjustment module, a howling suppression module, and a sound effect optimization feedback module that are communicatively connected in sequence; The audio signal acquisition module is used to collect the audio signal data of the sound system in real time through a sensor network, and the audio signal data includes spectrum, amplitude, phase, sound pressure level, and reverberation time; The signal analysis module is used to perform spectral feature decomposition on the audio signal data based on the non-negative matrix factorization algorithm, analyze the spectral features using the generative adversarial network algorithm, compare with the standard spectral curve, and generate a signal analysis report; The adaptive adjustment module is used to automatically adjust the gain, frequency response, and equalization parameters of the audio signal based on the signal analysis report using the fuzzy logic control algorithm; The howling suppression module is used to detect the howling noise component in the audio signal in real time using the adaptive feedback cancellation algorithm and perform dynamic suppression; The sound effect optimization feedback module is used to optimize the frequency, phase, and spatial positioning of the audio signal by constructing a high-order Bessel curve model of multi-channel audio based on the spatial sound effect enhancement algorithm based on high-order Bessel curve fitting; Among them, the operation process of the signal analysis module includes the following steps: Receive the audio signal data and perform preprocessing operations; Use the non-negative matrix factorization algorithm to perform spectral feature decomposition on the preprocessed audio signal data, generate multiple independent spectral bases and corresponding weight matrices, extract spectral feature information in different frequency bands, and mark the main frequency components of the signal; Adopt the generative adversarial network algorithm to perform multi-layer comparison of the spectral features with the standard spectral curve built in the system, generate a high-precision feature matching model through adversarial training, identify and highlight abnormal spectral components, and generate stability and similarity scores of the spectral features at the same time; Import the analysis results into the dynamic adjustment model, output a feature difference map based on the spectral feature report generated by the generative adversarial network algorithm, and generate a dynamic adjustment prompt according to the spectral difference and stability score; Integrate the comparison results and dynamic adjustment detection information to generate a signal analysis report, including detailed spectral component change trends, abnormal frequency identification results, gain distribution suggestions, frequency response curve matching degrees, and recommended adjustment ranges for each audio parameter.

2. The digital audio processing system for an audio amplifier according to claim 1, wherein The construction process of the non-negative matrix factorization algorithm includes the following steps: Divide the audio signal data into several adaptive sub-bands through an intelligent frequency band division algorithm, extract the main spectral features according to the characteristics of different frequency bands, and form an initial feature matrix; Use a sparsity control mechanism to initialize the basis matrix, and adjust the structure of the basis matrix through a unique sparse optimization function; In the decomposition iteration, dynamically adjust the sparse constraint parameter and the residual suppression weight to control the noise influence in the decomposition process, and gradually optimize the sparse structure of the spectral basis matrix; Integrate and output the final basis matrix and weight matrix, form a complete spectral feature set through multi-band feature summary, and generate a differential feedback of the spectral features.

3. A digital audio processing system for an audio device according to claim 1, characterized in that, The formula of the fuzzy logic control algorithm is as follows: where, ΔP represents the optimal gain adjustment amount; α and β respectively represent the weights of the spectral feature and the equalization feature in the overall adjustment; n represents the number of fuzzy rules, corresponding to different frequency band characteristics; w i represents the weight of the i-th fuzzy rule; f i (x) represents the spectral feature calculated according to the input variable x; g(x) represents the spectral feature weight function; a i and b i represent the integration interval; m represents the number of equalization features; h j (y j ) represents the equalization feature calculated based on the input variable y j .

4. A digital audio processing system for an audio device according to claim 1, characterized in that, The operation process of the howling suppression module includes the following steps: Real-time detect the howling components in the audio signal through the adaptive feedback cancellation algorithm, identify and locate the main frequencies causing howling, and at the same time construct a feedback path model; Utilize the feedback path model, apply an adaptive filter to generate a cancellation signal that is out of phase with the howling frequency, and dynamically eliminate the howling noise components by adjusting the filter weight coefficients in real time; According to the change of the howling frequency, adaptively adjust the gain and frequency band range of the filter, and continuously optimize the feedback path model through the adaptive feedback cancellation algorithm to enhance the suppression effect; After completing the feedback suppression, generate a howling suppression report, including the change trend of the howling frequency, the stability score of the feedback path, the suppression effect of each frequency band, and optimization suggestions for the feedback path model. The formula of the feedback path model is as follows:

5. A digital audio processing system for an audio device according to claim 4, characterized in that, The operation process of the sound effect optimization feedback module includes the following steps: Among them, y(t) represents the output audio signal after suppressing howling; x(t) represents the input original audio signal; N represents the number of feedback paths; G i (t) represents the adaptive gain control coefficient of the I-th feedback path; a i represents the influence degree of different paths on the feedback signal; τ i represents the delay time of the i-th feedback path; θ represents the suppression intensity coefficient; γ represents the amplitude power exponent of the feedback signal; f i (t) represents the amplitude of the feedback signal in the i-th feedback path.

6. A digital audio processing system for an audio device according to claim 1, wherein Collect the spectrum, phase and spatial positioning data of the audio through a multi-channel sensor, use the high-order Bessel curve fitting algorithm to generate a three-dimensional spatial sound effect model, and convert the spatial distribution characteristics of the sound source into a multi-order control point matrix; Based on the control point matrix generated by the modeling, fine-tune the frequency response and phase of each frequency band; Utilize the multi-channel feedback signal to adaptively regulate the Bessel curve control parameters and dynamically adjust the spatial distribution of the audio signal; Generate an optimization feedback report according to the result of the sound effect optimization, including the gain distribution map of the enhanced spatial sound effect, the error analysis of each sound source positioning, the optimized frequency response curve, and the recommended parameter adjustment range. The formula of the three-dimensional spatial sound effect model is as follows:

7. A digital audio processing system for an audio device according to claim 6, characterized in that, The system is applied to a digital audio processing system for a sound system as described in any one of claims 1-7, including the following steps: Among them, P(t, f, φ) represents the positioning point of the audio signal in three-dimensional space at time t, frequency f, and phase φ; B i，n (t) represents the i-th basis function of the Bessel curve; C i represents the coordinate matrix of the i-th control point of the Bessel curve; f i represents the frequency parameter of the i-th frequency band; φ i represents the phase offset of the i-th frequency band; μ represents the attenuation coefficient, controlling the influence of the distance d i (t) on the audio signal; d i (t) represents the dynamic distance function of the control point C i at time t; t represents the time parameter; n represents the order of the Bessel curve.

8. A digital audio processing method for an audio device, characterized in that, Real-time collect the audio signal data of the sound system through a sensor network; Based on the non-negative matrix factorization algorithm, perform spectral feature decomposition on the audio signal data, and use the generative adversarial network algorithm to analyze the spectral features, compare with the standard spectral curve, and generate a signal analysis report; Based on the signal analysis report, use the fuzzy logic control algorithm to automatically adjust the gain, frequency response and equalization parameters of the audio signal; Adopt the adaptive feedback cancellation algorithm to real-time detect the howling noise components in the audio signal and perform dynamic suppression; Based on the spatial sound effect enhancement algorithm based on high-order Bessel curve fitting, optimize the frequency, phase and spatial positioning of the audio signal by constructing a high-order Bessel curve model for multi-channel audio; Among them, the step of performing spectral feature decomposition on the audio signal data based on the non-negative matrix factorization algorithm, using the generative adversarial network algorithm to analyze the spectral features, comparing with the standard spectral curve, and generating a signal analysis report includes the following steps: Receive the audio signal data and perform preprocessing operations; Use the non-negative matrix factorization algorithm to perform spectral feature decomposition on the preprocessed audio signal data, generate multiple independent spectral bases and corresponding weight matrices, extract the spectral feature information of different frequency bands, and mark the main frequency components of the signal; ​ Using the generative adversarial network algorithm, the spectral features are compared with the standard spectral curves built into the system in multiple layers. Through adversarial training, a high-precision feature matching model is generated to identify and highlight abnormal spectral components, and at the same time, stability and similarity scores of the spectral features are generated; The analysis results are imported into the dynamic adjustment model. Based on the spectral feature report generated by the generative adversarial network algorithm, a feature difference map is output, and dynamic adjustment prompts are generated according to the spectral differences and stability scores; By comprehensively comparing the results and the dynamic adjustment detection information, a signal analysis report is generated, which includes the detailed change trend of spectral components, the identification results of abnormal frequencies, the gain distribution suggestions, the frequency response curve matching degree, and the recommended adjustment ranges for each audio parameter.

Images