Virtual bass enhancement method, system and device and storage medium

Abstract

The invention relates to a virtual bass enhancement method, which comprises the following steps: preprocessing an input audio signal, the preprocessing comprising sampling rate conversion, de-noising processing and volume standardization; extracting low-frequency components in the input audio signal through a low-pass filter; performing time domain analysis on the extracted low-frequency components, and generating odd harmonics and even harmonics of the low-frequency components by using a harmonic generation algorithm; performing low-frequency component identification on the preprocessed audio signal by using a support vector machine model so as to distinguish a low-frequency part needing to be enhanced; extracting a high-frequency component in the input audio signal through a high-pass filter; and carrying out superposition synthesis on the generated harmonic signal, the identified low-frequency component and the extracted high-frequency component according to a preset proportion to obtain an enhanced low-frequency audio output signal. The method can solve the problems that multiple virtual bass algorithms exist in the market, but most of the virtual bass algorithms are high in processing complexity, poor in bass reduction effect or high in cost.

Description

Technical Field

[0001] The present invention relates to the field of digital audio, in particular to a virtual bass enhancement method, system, device and storage medium. Background Art

[0002] Micro speakers are small in size and have limited power, so their low-frequency response is generally weak, which makes the sound thin and lacking in power when playing audio with a lot of low-frequency components. Traditional solutions include using external speakers or audio processing software to enhance the bass effect, but these methods either increase the size and weight of the device or make it difficult to achieve the desired bass enhancement effect.

[0003] As an emerging audio processing technology, virtual bass technology can significantly improve the low-frequency performance of audio without changing the hardware structure by simulating bass effects instead of actually generating bass. Currently, there are many virtual bass algorithms on the market, but most of them have problems such as high processing complexity, poor bass restoration effect or high cost. Summary of the invention

[0004] The embodiments of the present application provide a virtual bass enhancement method, system, device and storage medium, which can solve the problems that there are many virtual bass algorithms on the market, but most of them have high processing complexity, poor bass restoration effect or high cost.

[0005] According to one aspect of the present disclosure, a virtual bass enhancement method is provided, comprising:

[0006] Preprocessing the input audio signal, wherein the preprocessing includes sampling rate conversion, noise removal and volume normalization;

[0007] Extract low-frequency components from the input audio signal through a low-pass filter;

[0008] Performing time domain analysis on the extracted low-frequency components, and using a harmonic generation algorithm to generate odd harmonics and even harmonics of the low-frequency components;

[0009] Use the support vector machine model to identify the low-frequency components of the pre-processed audio signal to distinguish the low-frequency parts that need to be enhanced;

[0010] Extract high-frequency components from the input audio signal through a high-pass filter;

[0011] The generated harmonic signal, the identified low-frequency component and the extracted high-frequency component are superimposed and synthesized according to a preset ratio to obtain an audio output signal with enhanced low frequency.

[0012] Optionally, the harmonics include at least a first harmonic, a second harmonic, a third harmonic and a fourth harmonic.

[0013] Optionally, the harmonic generation algorithm calculates the proportionality coefficient and phase information of each harmonic according to the time domain waveform characteristics of the low-frequency component to generate a corresponding harmonic signal.

[0014] Optionally, during the training process, the support vector machine model is trained using a data set containing low-frequency signal samples and non-low-frequency signal samples to achieve accurate recognition of low-frequency components in the audio signal.

[0015] Optionally, the signal superposition step includes phase calibration and amplitude balancing to ensure the naturalness and distortion-freeness of the final output signal.

[0016] According to another aspect of the present application, a virtual bass enhancement system is provided, comprising:

[0017] A signal preprocessing module, used for preprocessing the input audio signal;

[0018] A low-frequency component extraction and harmonic generation module is used to extract low-frequency components and generate corresponding harmonic signals;

[0019] The low-frequency component recognition module uses a support vector machine model to recognize the low-frequency components of the preprocessed audio signal;

[0020] A high-frequency component extraction module is used to extract high-frequency components in audio signals;

[0021] The signal synthesis module is used to superimpose and synthesize the generated harmonic signal, the identified low-frequency component and the extracted high-frequency component to obtain an audio output signal with enhanced low frequency.

[0022] Optionally, the low-frequency component extraction and harmonic generation module further includes a time domain analysis unit and a harmonic generation unit, the time domain analysis unit is used to analyze the time domain waveform characteristics of the low-frequency component, and the harmonic generation unit generates a corresponding harmonic signal according to the analysis result.

[0023] According to another aspect of the present application, an electronic device is provided, including a processor and a memory, wherein a computer program is stored in the memory, and when the processor executes the program, the virtual bass enhancement method described in any one of the aforementioned embodiments is implemented.

[0024] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the virtual bass enhancement method described in any one of the aforementioned embodiments is implemented.

[0025] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0026] The low-frequency components in the audio signal are accurately extracted through a low-pass filter, and its harmonics are generated using a harmonic generation algorithm, effectively making up for the shortcomings of the micro speaker in the low-frequency band. At the same time, the introduction of the support vector machine model enables the system to intelligently identify and enhance the low-frequency parts that really need to be enhanced, avoiding unnecessary processing, thereby significantly improving the sound quality of the audio signal in the low-frequency band. Users can feel a fuller and deeper bass effect, which improves the overall listening experience.

[0027] While enhancing the low frequencies, the system extracts and retains the high frequency components in the audio signal through a high-pass filter, ensuring that the sound quality of the high frequency part is not affected. This processing method avoids the high frequency distortion problem that may occur in traditional bass enhancement methods. The high frequency part of the audio signal remains clear and bright, complementing the enhanced low frequency part, and the overall sound quality is more balanced.

[0028] Since the algorithm is designed based on the basic principles of signal processing and does not rely on a specific hardware platform or audio format, it has strong adaptability and a wide range of applications. Whether it is a portable device such as a mobile phone or a tablet, or a large-size audio system such as a car audio or a home theater, this technical solution can be used for bass enhancement. The versatility and flexibility of the technical solution enable it to meet the needs of improving sound quality in different scenarios and has broad market prospects.

[0029] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0031] Figure 1 is a flow chart of a virtual bass enhancement method shown in an exemplary embodiment of the present application;

[0032] Figure 2 It is a schematic diagram of a virtual bass enhancement system shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0033] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0034] Figure 1A flowchart of a virtual bass enhancement method according to an exemplary embodiment of the present application is shown. The workflow may include the following steps:

[0035] S101 pre-processes the input audio signal, wherein the pre-processing includes sampling rate conversion, noise removal and volume normalization.

[0036] At the beginning of the algorithm, the input audio signal needs to be preprocessed. The main purpose of this step is to provide a stable and standardized input for subsequent signal processing. The preprocessing process may include signal sampling rate conversion, denoising, and possible volume normalization.

[0037] The sampling rate conversion can convert the sampling rate of the input signal to a suitable value according to the needs of the algorithm to ensure the efficiency and accuracy of the algorithm processing. For example, the sampling rate of the input signal can be converted to a common audio sampling rate such as 44.1kHz or 48kHz.

[0038] De-noising can use digital filtering technology, such as Wiener filtering or Kalman filtering, to denoise the input signal to reduce the impact of background noise on subsequent signal processing.

[0039] Volume normalization adjusts the volume of the input signal to a uniform level to ensure that the algorithm maintains consistent performance at different volume levels.

[0040] S102 extracts low-frequency components from the input audio signal through a low-pass filter.

[0041] After the preprocessing is completed, the next step is to extract the low-frequency components of the input signal and generate harmonics. This step is one of the core of the algorithm and is directly related to the low-frequency sound quality of the final output signal.

[0042] The input signal is filtered using a low-pass filter (LPF) to extract its low-frequency components. The cutoff frequency of the low-pass filter should be determined according to the physical characteristics of the micro speaker, and a value below the speaker cutoff frequency is usually selected as the cutoff frequency.

[0043] S103 performs time domain analysis on the extracted low-frequency components, and generates odd harmonics and even harmonics of the low-frequency components using a harmonic generation algorithm.

[0044] The extracted low-frequency components are analyzed in the time domain to identify their waveform characteristics and frequency components. This step may involve signal processing techniques such as signal period detection and frequency estimation.

[0045] According to the results of time domain low-frequency analysis, the number of odd harmonics and even harmonics of the low-frequency component and the proportional coefficient of each harmonic are determined by using the harmonic generator. The harmonic signal of its low-frequency component is obtained, and the mathematical expression can be expressed as:

[0046] y(n)=α1x(n)+α2x 2 (n)+α3x 3 (n)+α4x 4 (n)+…

[0047] Generally, the higher the harmonic order, the less information it contains. In this patent, the maximum value of the harmonic order is 4, which includes the first harmonic, the second harmonic, the third harmonic and the fourth harmonic. α1, α2, α3, and α4 represent the proportional coefficients of the first harmonic, the second harmonic, the third harmonic and the fourth harmonic respectively.

[0048] S104 uses a support vector machine model to identify low-frequency components of the preprocessed audio signal to distinguish the low-frequency part that needs to be enhanced.

[0049] The low-frequency component recognition module classifies and outputs the low-frequency components in the input signal by using a support vector machine (SVM) model. The specific process includes the following steps:

[0050] Signal framing: The input signal is framed according to a certain frame length and frame shift. In this patent, the frame length is set to 1024 sampling points and the frame shift is 50%, that is, 512 sampling points. This ensures that each frame of the signal contains enough low-frequency information for the SVM model to analyze.

[0051] Window function processing: In order to reduce the spectrum leakage problem caused by frame processing, it is necessary to apply a window function to each frame signal for weighted processing. In this patent, the Hamming window is selected as the window function, and its calculation formula is as follows:

[0052] w(n)=0.54-0.46cos[2πn / (L-1)], 0≤n≤L-1

[0053] Where w(n) is the value of the window function at position n, n is the index of the sample point in the window (usually starting from 0), and L is the length of the window (i.e. the total number of sample points).

[0054] Feature extraction: Perform a fast Fourier transform (FFT) on each frame of the signal to obtain its spectrum representation. Then calculate the energy of the spectrum below 800Hz as the feature vector for the input of the SVM model. The calculation formula for signal energy is as follows:

[0055] E(x)=[x(n)]^2

[0056] X(n) represents the value of signal x at sample n

[0057] SVM model training and classification: Use the pre-trained SVM model to classify the feature vector and determine whether it is a low-frequency signal. Its mathematical expression can be expressed as:

[0058] F(x)=w*x+b, where w is the weight vector, x is the feature vector, and b is the bias vector.

[0059] S105 extracts high-frequency components from the input audio signal through a high-pass filter.

[0060] After the low-frequency component is identified, the next step is to extract the high-frequency component of the input signal. The input signal is filtered using a high-pass filter to extract its high-frequency components. This step is similar to the low-frequency component extraction, but a high-pass filter (HPF) is used.

[0061] S106: The generated harmonic signal, the identified low-frequency component and the extracted high-frequency component are superimposed and synthesized according to a preset ratio to obtain an audio output signal with enhanced low frequency.

[0062] The harmonic signal generated after low-pass filtering, the high-frequency signal in the original signal, and the identified low-frequency components are superimposed and synthesized to obtain the final output signal.

[0063] The harmonic signal, high-frequency signal and low-frequency component are superimposed in a certain proportion. The superposition ratio can be adjusted according to the specific application scenario and auditory effect.

[0064] When superimposed, adjustments are made based on the relative strength and phase relationship of each component to ensure the sound quality and harmony of the final output signal. After the above processing, the output signal is significantly enhanced in the low-frequency part while maintaining the high-frequency components and timbre characteristics of the original signal.

[0065] See also Figure 2 In a preferred embodiment, a virtual bass enhancement system 201 is also provided, comprising:

[0066] The signal preprocessing module 202 is used to preprocess the input audio signal;

[0067] A low-frequency component extraction and harmonic generation module 203 is used to extract low-frequency components and generate corresponding harmonic signals;

[0068] A low frequency component identification module 204 uses a support vector machine model to identify low frequency components of the pre-processed audio signal;

[0069] A high frequency component extraction module 205, used to extract high frequency components in the audio signal;

[0070] The signal synthesis module 206 is used to superimpose and synthesize the generated harmonic signal, the identified low-frequency component and the extracted high-frequency component to obtain an audio output signal with enhanced low frequency.

[0071] According to another aspect of the present application, an electronic device is provided, including a processor and a memory, wherein a computer program is stored in the memory, and when the processor executes the program, the virtual bass enhancement method described in any one of the aforementioned embodiments is implemented.

[0072] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the virtual bass enhancement method described in any one of the aforementioned embodiments is implemented.

[0073] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0074] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A virtual bass enhancement method, characterized in that: include: Preprocessing the input audio signal, wherein the preprocessing includes sampling rate conversion, noise removal and volume normalization; Extract low-frequency components from the input audio signal through a low-pass filter; Performing time domain analysis on the extracted low-frequency components, and using a harmonic generation algorithm to generate odd harmonics and even harmonics of the low-frequency components; Use the support vector machine model to identify the low-frequency components of the pre-processed audio signal to distinguish the low-frequency parts that need to be enhanced; Extract high-frequency components from the input audio signal through a high-pass filter; The generated harmonic signal, the identified low-frequency component and the extracted high-frequency component are superimposed and synthesized according to a preset ratio to obtain an audio output signal with enhanced low frequency.

2. The method according to claim 1, characterized in that The harmonics include at least first harmonic, second harmonic, third harmonic and fourth harmonic.

3. The method according to claim 1, characterized in that The harmonic generation algorithm calculates the proportional coefficient and phase information of each harmonic according to the time domain waveform characteristics of the low-frequency component to generate a corresponding harmonic signal.

4. The method according to claim 1, characterized in that: During the training process, the support vector machine model is trained using a data set containing low-frequency signal samples and non-low-frequency signal samples to achieve accurate recognition of low-frequency components in audio signals.

5. The method according to claim 1, characterized in that The signal summing step includes phase calibration and amplitude balancing to ensure the naturalness and distortion-free of the final output signal.

6. A virtual bass enhancement system, characterized in that: include: A signal preprocessing module, used for preprocessing the input audio signal; A low-frequency component extraction and harmonic generation module is used to extract low-frequency components and generate corresponding harmonic signals; The low-frequency component recognition module uses a support vector machine model to recognize the low-frequency components of the preprocessed audio signal; A high-frequency component extraction module is used to extract high-frequency components in audio signals; The signal synthesis module is used to superimpose and synthesize the generated harmonic signal, the identified low-frequency component and the extracted high-frequency component to obtain an audio output signal with enhanced low frequency.

7. The virtual bass enhancement system according to claim 6, characterized in that: The low-frequency component extraction and harmonic generation module further includes a time domain analysis unit and a harmonic generation unit. The time domain analysis unit is used to analyze the time domain waveform characteristics of the low-frequency component, and the harmonic generation unit generates a corresponding harmonic signal according to the analysis result.

8. An electronic device, comprising a processor and a memory, wherein a computer program is stored in the memory, and when the processor executes the program, the virtual bass enhancement method according to any one of claims 1 to 5 is implemented.

9. A storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the virtual bass enhancement method according to any one of claims 1 to 5 is implemented.

Images