Vehicle noise reduction method, controller, vehicle noise reduction system, vehicle, storage medium and program product

Abstract

The present disclosure relates to the technical field of vehicle noise reduction, and provides a vehicle noise reduction method, controller, vehicle noise reduction system, vehicle, storage medium, and program product. The vehicle noise reduction method includes: obtaining a first audio signal collected by a microphone in the vehicle, the first audio signal being used to determine a second audio signal at a target location in the vehicle, and the second audio signal being used to generate an inverted audio signal; and controlling a speaker to play the inverted audio signal so that the inverted audio signal can be used to reduce the second audio signal when it propagates to the target location. The inverted audio signal generated by the vehicle noise reduction method can be used to reduce the second audio signal when it propagates to the target location, directly and accurately achieving noise cancellation at the target location and improving the noise reduction effect.

Description

Technical Field

[0001] The present disclosure relates to the technical field of vehicle noise reduction, and in particular, to a vehicle noise reduction method, a controller, a vehicle noise reduction system, a vehicle, a storage medium, and a program product. Background Art

[0002] Active noise cancellation (ANC) uses electronic systems to generate signals that are opposite in phase to ambient noise, canceling or reducing the effects of the noise through signal interaction. The core principle of ANC is based on the destructive interference phenomenon of superimposed sound waves. Specifically, when two sound waves with the same frequency and amplitude but opposite phases meet, they cancel each other out, achieving the desired noise reduction effect.

[0003] In related technologies, vehicle noise reduction mainly relies on microphones to collect noise in real time, generate anti-phase sound waves through algorithms, and then emit noise reduction through speakers. However, because the anti-phase sound waves are generated based on the noise at the microphone position, their phase and intensity matching is only optimal at that location. When propagating to other areas in the car, the destructive interference effect is greatly weakened by the attenuation of sound waves, phase changes, and the complex acoustic environment inside the car. Therefore, the generated anti-phase sound waves have a better noise reduction effect near the microphone, but the noise reduction effect is poor in other locations in the car. Summary of the Invention

[0004] The purpose of the present disclosure is to provide a vehicle noise reduction method, a controller, a vehicle noise reduction system, a vehicle, a storage medium and a program product to solve the technical problems in the related art.

[0005] In order to achieve the above objectives, the present disclosure provides a vehicle noise reduction method, the vehicle noise reduction method comprising:

[0006] Acquire a first audio signal collected by a microphone in the vehicle, wherein the first audio signal is used to determine a second audio signal at a target position in the vehicle, and the second audio signal is used to generate an inverted audio signal;

[0007] The loudspeaker is controlled to play the inverted audio signal, so that the inverted audio signal can be used to reduce the second audio signal when it propagates to the target location.

[0008] Optionally, the inverted audio signal includes a first inverted audio signal obtained by inverting the second audio signal, and a second inverted audio signal determined according to a filtering parameter and the first audio signal.

[0009] Optionally, the filtering parameters are generated based on a second secondary acoustic feedback signal and a second noise audio signal, wherein the second noise audio signal represents an audio signal formed when the noise source propagates to the target position, and the second secondary acoustic feedback signal is an audio signal formed when the inverted audio signal played by the speaker is transmitted to the target position.

[0010] Optionally, the second noise audio signal is generated according to the first audio signal and a first secondary acoustic feedback signal, where the first secondary acoustic feedback signal is an audio signal formed when an inverted audio signal played by the loudspeaker is transmitted to the microphone.

[0011] Optionally, the vehicle noise reduction method further includes:

[0012] Determine a first noise audio signal based on the first audio signal and the first secondary acoustic feedback signal, wherein the first noise audio signal represents an audio signal formed when a noise source propagates to the microphone;

[0013] The second noise audio signal is determined based on the first noise audio signal and a preset transfer relationship, wherein the preset transfer relationship represents a correspondence between the audio signal at the microphone and the audio signal at the target position.

[0014] Optionally, the first secondary acoustic feedback signal is determined based on the inverted audio signal played by the speaker and a pre-calibrated first transfer function, wherein the first transfer function represents the acoustic path characteristics of the audio signal played by the speaker to the microphone; and / or,

[0015] The second secondary acoustic feedback signal is determined based on the inverted audio signal played by the loudspeaker and a pre-calibrated second transfer function, wherein the second transfer function represents the acoustic path characteristics of the audio signal played by the loudspeaker propagating to the target position.

[0016] Optionally, the first audio signal includes a plurality of sub-band signals in different frequency ranges, and the vehicle noise reduction method further includes:

[0017] For each of the sub-band signals, a second inverse phase audio signal corresponding to the sub-band signal is determined according to the filtering parameters and the sub-band signal.

[0018] Optionally, the second audio signal is determined according to the first audio signal and a pre-trained processing model.

[0019] Optionally, the second audio signal is determined according to the first audio signal, air-conditioning operating parameters when the microphone collects the first audio signal, and a pre-trained processing model.

[0020] Optionally, the target position is the ear position of a passenger in the vehicle.

[0021] The present disclosure also provides a controller, comprising:

[0022] a memory having a computer program stored thereon;

[0023] A processor is configured to execute the computer program in the memory to implement any one of the steps of the above-mentioned vehicle noise reduction method.

[0024] The present disclosure further provides a vehicle noise reduction system, the vehicle noise reduction system comprising:

[0025] microphone;

[0026] speaker;

[0027] The above-mentioned controller, the microphone and the speaker are all connected to the controller.

[0028] The present disclosure also provides a vehicle, comprising the above-mentioned controller or the above-mentioned vehicle noise reduction system.

[0029] The present disclosure further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of any one of the above-mentioned vehicle noise reduction methods are implemented.

[0030] The present disclosure further provides a computer program product, comprising a computer program, which implements the steps of any one of the above-mentioned vehicle noise reduction methods when executed by a processor.

[0031] By adopting the above technical solution, the second audio signal at the target position in the car (such as the position of the passenger's ear in the car) is accurately determined based on the first audio signal collected by the microphone; then an optimized inverted audio signal is generated for the second audio signal, and then the inverted audio signal is played by controlling the speaker. When the inverted audio signal is transmitted to the target position, it can be used to reduce the second audio signal, directly and accurately achieving noise cancellation at the target position, thereby improving the noise reduction effect.

[0032] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following detailed description, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0034] Figure 1 The figure is a flow chart showing a vehicle noise reduction method according to an exemplary embodiment.

[0035] Figure 2 The figure is a schematic diagram showing the relationship of audio signals according to an exemplary embodiment.

[0036] Figure 3 The figure is a schematic diagram showing filtering and segmenting of a first audio signal according to an exemplary embodiment.

[0037] Figure 4 The figure is a schematic diagram showing a transmission path of air-conditioning noise according to an exemplary embodiment.

[0038] Figure 5 The figure is a schematic diagram showing training of a modeling network according to an exemplary embodiment.

[0039] Figure 6 The figure is a schematic diagram showing an application of a modeling network according to an exemplary embodiment.

[0040] Figure 7 The figure is a schematic diagram showing a vehicle noise reduction method according to an exemplary embodiment.

[0041] Figure 8 The figure is a block diagram of a vehicle noise reduction device according to an exemplary embodiment. DETAILED DESCRIPTION

[0042] The following describes the specific embodiments of the present disclosure in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure and are not intended to limit the present disclosure.

[0043] In the following description, words such as “first” and “second” are only used for the purpose of distinguishing the description and should not be understood as indicating or implying relative importance or order.

[0044] ANC is a technology that uses electronic systems to generate signals that are in phase with the ambient noise, canceling or reducing the effects of the noise through the interaction of these signals. The core principle of active noise cancellation is based on the destructive interference phenomenon of superimposed sound waves. This is because when two sound waves of the same frequency and amplitude but opposite phases meet, they cancel each other out, achieving the desired noise reduction effect.

[0045] In related technologies, vehicle noise reduction mainly relies on microphones to collect noise in real time, generate anti-phase sound waves through algorithms, and then emit noise reduction through speakers. However, because the anti-phase sound waves are generated based on the noise at the microphone position, their phase and intensity matching is only optimal at that location. When propagating to other areas in the car, the destructive interference effect is greatly weakened by the attenuation of sound waves, phase changes, and the complex acoustic environment inside the car. Therefore, the generated anti-phase sound waves have a better noise reduction effect near the microphone, but the noise reduction effect is poor in other locations in the car.

[0046] Patent CN111572311A, titled "Active Noise Reduction System for Electric Vehicles and Electric Vehicles," discloses a noise reduction method and system for addressing fan noise in electric vehicle air conditioners. The system comprises a main noise collection device, an ear positioning collection device, and a signal processor. The main noise collection device is located within the electric vehicle's air conditioning compartment, while the ear positioning collection device is located inside the vehicle. Both the main noise collection device and the ear positioning collection device are communicatively connected to the signal processor, which receives noise signals and binaural position signals and updates the output phase sound wave signal in real time. The electric vehicle's onboard audio and video speakers receive the phase sound wave signal and output an anti-phase sound wave, thereby reducing the air conditioning compartment fan noise.

[0047] The above solution has the following defects:

[0048] First, since the automobile air conditioning ducts are relatively complex and there are many air outlets in the cabin, according to the above patent, "the main noise collection device is set in the air conditioning box of the electric vehicle", it is necessary to arrange more reference microphones, which will cause many hardware problems: (1) The hardware system is relatively complex; (2) The high price of microphones will result in high system costs; (3) There are too many wire harnesses between microphones, which increases the difficulty of hardware system assembly; (4) The controller needs to receive more microphone signals, so the controller section needs to be equipped with more signal receiving modules;

[0049] Second, deploying more reference microphones will cause many software problems: (1) The controller needs to process more input signals, so the underlying driver design will be more complicated; (2) A complex filter system needs to be designed from the reference microphone to the speaker, which will occupy a higher system computing power requirement.

[0050] In order to solve the above technical problems, a second audio signal at a target position in the vehicle (such as the position of a passenger's ear) is accurately determined based on the first audio signal collected by a microphone; then an optimized inverted audio signal is generated for the second audio signal, and the inverted audio signal is played by controlling the speaker. When the inverted audio signal is transmitted to the target position, it can be used to reduce the second audio signal, thereby directly and accurately achieving noise cancellation at the target position and improving the noise reduction effect.

[0051] Figure 1 This is a flow chart showing a vehicle noise reduction method according to an exemplary embodiment. The vehicle noise reduction method can be applied to a controller on a vehicle or a vehicle noise reduction system on a vehicle. Figure 1 , the vehicle noise reduction method may include step S1 and step S2.

[0052] Step S1: obtaining a first audio signal collected by a microphone in the vehicle, wherein the first audio signal is used to determine a second audio signal at a target position in the vehicle, and the second audio signal is used to generate an inverted audio signal.

[0053] The microphone is used to collect sound waves from inside the vehicle and convert them into electrical signals. The microphone can be installed anywhere in the vehicle, such as the center console, instrument panel, or A-pillar. The microphone can reuse the voice wake-up microphone in the cabin.

[0054] The first audio signal may be a mixed audio signal of an audio signal formed by an audio signal played by a speaker and propagated to the microphone, and an audio signal formed by a noise source and propagated to the microphone.

[0055] The target position in the car is the position in the car where noise reduction is required. The target position can be, but is not limited to, the position of the passenger's head in the car, the position of the passenger's ear in the car, the driver's seat position, the co-pilot seat position, the rear seat position, etc.

[0056] The second audio signal may be a mixed audio signal of an audio signal formed by an audio signal played by a speaker and propagated to the target position, and an audio signal formed by an audio signal propagated from a noise source to the target position.

[0057] The anti-phase audio signal may be an audio signal for suppressing the second audio signal.

[0058] Step S2: Control the loudspeaker to play the inverted audio signal, so that the inverted audio signal can be used to reduce the second audio signal when it propagates to the target location.

[0059] First, the audio signal at the target position serving as the noise receiving end (for example, the ear of a passenger in a car), i.e., the second audio signal, is determined. Then, based on the second audio signal "heard" at the target position, an inverted audio signal for suppression is generated. The speaker is controlled to play the inverted audio signal, so that when the inverted audio signal is transmitted to the target position, it can be used to reduce the second audio signal, thereby accurately reducing noise at the target position.

[0060] Based on the first audio signal collected by the microphone, the second audio signal at the target position in the car (such as the position of the passenger's ear in the car) is accurately determined; then an optimized inverted audio signal is generated for the second audio signal, and then the inverted audio signal is played by controlling the speaker. When the inverted audio signal is transmitted to the target position, it can be used to reduce the second audio signal, directly and accurately achieving noise cancellation at the target position, thereby improving the noise reduction effect.

[0061] In one possible implementation, see Figure 2 The inverted audio signal includes a first inverted audio signal obtained by inverting the second audio signal, and a second inverted audio signal determined according to the filtering parameter and the first audio signal.

[0062] After the first inverted phase audio signal is played, it will be recorded by the microphone and become part of the first audio signal. Therefore, directly playing the first inverted phase audio signal will result in errors. By filtering the parameters and the first audio signal, a second inverted phase audio signal can be obtained to compensate for the first inverted phase audio signal.

[0063] An inversion operation is performed on the second audio signal, that is, its phase is changed by 180 degrees, to obtain a first inverted audio signal.

[0064] The process of determining the second inverted audio signal is as follows:

[0065] Determine a second noise audio signal based on the first audio signal, wherein the second noise audio signal represents an audio signal formed when the noise source propagates to the target position;

[0066] A second inverse phase audio signal is determined based on the second noise audio signal and the filter parameters.

[0067] In a possible implementation, the second noise audio signal is generated according to the first audio signal and a first secondary acoustic feedback signal, where the first secondary acoustic feedback signal is an audio signal formed when an inverted audio signal played by a loudspeaker is transmitted to a microphone.

[0068] In a possible implementation, the vehicle noise reduction method may further include:

[0069] Determine a first noise audio signal based on the first audio signal and the first secondary acoustic feedback signal, wherein the first noise audio signal represents an audio signal formed when a noise source propagates to a microphone;

[0070] A second noise audio signal is determined based on the first noise audio signal and a preset transfer relationship, wherein the preset transfer relationship represents a correspondence between the audio signal at the microphone and the audio signal at the target position.

[0071] It should be understood that, during this process, the current inverted audio signal has not yet been generated, so the first secondary acoustic feedback signal is the audio signal formed when the last determined inverted audio signal played by the loudspeaker is propagated to the microphone.

[0072] In one embodiment, in a scenario where in-vehicle air conditioning noise is reduced, the first noise audio signal is the air conditioning noise audio signal collected at the microphone. The audio signal formed by the air conditioning noise propagating to the microphone is mixed with the audio signal formed by the inverted audio signal played by the speaker propagating to the microphone, resulting in a mixed audio signal formed at the microphone. Specifically, the first noise audio signal is mixed with the first secondary acoustic feedback signal to obtain the first audio signal. Therefore, the first noise audio signal can be obtained based on the first audio signal and the first secondary acoustic feedback signal.

[0073] The first noise audio signal can be obtained by subtracting the first audio signal from the first secondary acoustic feedback signal.

[0074] In a possible implementation, the first secondary acoustic feedback signal is determined based on an inverted audio signal played by a speaker and a pre-calibrated first transfer function, wherein the first transfer function characterizes acoustic path characteristics from the audio signal played by the speaker to the microphone.

[0075] A frequency sweep signal containing a noise reduction frequency band is emitted through a loudspeaker, corresponding to the audio signal picked up by the microphone. The first transfer function between the frequency sweep signal and the audio signal at the microphone can be calibrated based on algorithms such as the variable step-size least mean square algorithm (LMS).

[0076] The first secondary acoustic feedback signal is a convolution of the anti-phase audio signal and the first transfer function.

[0077] The preset transfer relationship may be, but is not limited to, a transfer relationship mapping table, a transfer relationship expression, a transfer relationship function, and the like.

[0078] The calibration process of the preset transfer relationship is as follows:

[0079] When the speaker is not playing, a noise source emits noise that propagates to the microphone and a target location. The microphone collects multiple first noise audio signals. A sound pickup device is provided at the target location, and the sound pickup device picks up multiple second noise audio signals. Based on the multiple first noise audio signals and the multiple second noise audio signals, a correspondence between the audio signal at the microphone and the audio signal at the target location can be determined, that is, a preset transfer relationship between the first noise audio signals and the second noise audio signals is calibrated. The sound pickup device provided at the target location can be a microphone.

[0080] Furthermore, to further improve the accuracy of the preset transfer relationship, the influence of air conditioning operating parameters on the preset transfer relationship can also be considered when reducing in-vehicle air conditioning noise. The preset transfer relationship is calibrated under different air conditioning operating parameters, such as different modes, wind speeds, temperature settings, wind directions, circulation, and other state parameters. Correspondingly, determining the second noise audio signal based on the first noise audio signal and the preset transfer relationship can be understood as determining the second noise audio signal generated at a target location within the vehicle based on the air conditioning operating parameters, the first noise audio signal, and the preset transfer relationship.

[0081] Among them, the modes can include cooling, heating, defrosting, ventilation and other modes; the wind speed can include different gear information, such as 1, 2, 3...7 and other gears; the temperature setting value can include different set temperatures, such as 18℃, 19℃...30℃, etc.; the wind direction can include blowing on the face, blowing on the feet, blowing on the face + blowing on the feet and other settings; the cycle can include internal circulation, external circulation, etc.

[0082] Determining the second inverted audio signal based on the second noise audio signal and the filtering parameters can be understood as filtering the second noise audio signal using the filtering parameters to obtain a filtered signal; then, inverting the filtered signal, i.e., changing its phase by 180 degrees, to obtain the second inverted audio signal.

[0083] In one possible implementation, the filtering parameters are generated based on a second secondary acoustic feedback signal and a second noise audio signal, wherein the second noise audio signal represents an audio signal formed when the noise source propagates to the target position, and the second secondary acoustic feedback signal is an audio signal formed when the inverted audio signal played by the speaker is transmitted to the target position.

[0084] The generation process of filter parameters is as follows:

[0085] generating an error signal based on the second secondary acoustic feedback signal and the second noise audio signal;

[0086] Based on the error signal and the second noisy audio signal, a filtering parameter is determined.

[0087] It should be understood that, during this process, the current inverted audio signal has not yet been generated, so the second secondary acoustic feedback signal is the audio signal formed when the last determined inverted audio signal played by the loudspeaker propagates to the target position.

[0088] For example, in a scenario where air conditioning noise in a vehicle is being reduced, the second noise audio signal can be the air conditioning noise audio signal generated at a target location. The audio signal generated by the air conditioning noise propagating to the target location is mixed with the audio signal generated by the inverted audio signal played by the speaker propagating to the target location, resulting in a mixed audio signal generated at the target location. Specifically, the second noise audio signal is mixed with the second secondary acoustic feedback signal to generate an error signal.

[0089] The error signal can be obtained by adding the second noise audio signal and the second secondary acoustic feedback signal.

[0090] In one embodiment, an adaptive filtering algorithm can be used, using the second noisy audio signal as the desired signal and the error signal as the input signal. The filter coefficients are continuously adjusted to minimize the error between the filter output and the desired signal. The filter coefficients ultimately converged are the desired filtering parameters. The adaptive filtering algorithm can include, but is not limited to, the LMS algorithm, the Recursive Least Squares (RLS) algorithm, the Affine Projection (AP) algorithm, and the like.

[0091] In a possible implementation, the second secondary acoustic feedback signal is determined based on the inverted audio signal played by the speaker and a pre-calibrated second transfer function, wherein the second transfer function characterizes the acoustic path characteristics of the audio signal played by the speaker propagating to the target position.

[0092] A sweep frequency signal containing a noise reduction frequency band is emitted through a loudspeaker, and a pickup device is set at the target position to pick up the audio signal accordingly. The calibration of the second transfer function between the sweep frequency signal and the audio signal at the target position can be completed based on algorithms such as LMS.

[0093] The second secondary acoustic feedback signal may be a convolution of the anti-phase audio signal and the second transfer function.

[0094] In a possible implementation, the first audio signal includes multiple sub-band signals in different frequency ranges, and the vehicle noise reduction method may further include:

[0095] For each sub-band signal, a second inverse phase audio signal corresponding to the sub-band signal is determined according to the filter parameter and the sub-band signal.

[0096] In order to improve the accuracy, a certain number of sub-band filters can be designed to filter and divide the first audio signal into multiple sub-band signals in different frequency ranges. Figure 3, sub-band signal 1 is separated out by filtering through sub-band filter 1, sub-band signal 2 is separated out by filtering through sub-band filter 2…, sub-band signal N is separated out by filtering through sub-band filter N. The specific number of segments and the bandwidth of each segment can be determined according to the actual noise situation, and the frequency band width of each segment can be set as needed, which can be set to equal width or non-equal width.

[0097] For example, if the frequency band of the air-conditioning noise to be processed is 0-1000 Hz, it can be divided into N parts (N is 20), that is, the first audio signal collected by the microphone is divided into sub-band signals with a bandwidth of 50 Hz.

[0098] For each subband signal, a second inverted audio signal corresponding to the subband signal is determined based on the filtering parameters and the subband signal. The same process for determining the second inverted audio signal based on the filtering parameters and the first audio signal can be referred to. By replacing the first audio signal with the subband signal, the second inverted audio signal corresponding to the subband signal can be obtained. This embodiment will not be described in detail here.

[0099] See also Figure 4 When the speakers are not playing and the air conditioner is turned on, the air conditioner noise (noise source), including blower noise, compressor noise, eddy current noise, and duct vibration noise, propagates to the microphone via transmission path ①, causing the microphone to pick up the noise audio signal. Similarly, the air conditioner noise propagates to the target location via transmission path ②. If a pickup device is placed at the target location, the pickup device will also pick up the noise audio signal. Since the air conditioner noise radiates from the instrument panel into the cabin, and the microphone is closer to the instrument panel, that is, transmission path ① is shorter than transmission path ②, using the noise audio signal picked up by the microphone to predict the noise audio signal formed at the target location is causal. Furthermore, since the air conditioner noise source is the same, the noise audio signal picked up by the microphone and the noise audio signal at the target location also meet the coherence requirement. Since the speaker is not playing, the microphone will pick up a noise audio signal that is equal to the first audio signal, and the sound pickup device will pick up a noise audio signal that is equal to the second audio signal. In this way, multiple microphones will pick up noise audio signals and corresponding multiple sound pickup devices will pick up noise audio signals, so that the mapping relationship between the first audio signal and the second audio signal can be calibrated. Then, based on the first audio signal collected by the microphone and the pre-calibrated mapping relationship, the second audio signal formed at the target position can be obtained.

[0100] In one embodiment, the second audio signal may be generated according to the first audio signal and a pre-calibrated mapping relationship.

[0101] In a possible implementation, the second audio signal is determined according to the first audio signal and a pre-trained processing model.

[0102] The first audio signal collected by the microphone is input into a pre-trained neural network to obtain a second audio signal.

[0103] The pre-trained neural network is used to predict the second audio signal formed at the target position based on the first audio signal collected by the microphone.

[0104] Pre-trained neural networks can be trained in the following ways:

[0105] Audio signals collected by a large number of microphones are collected, and the audio signals at the target position are collected as labels. They are trained through supervised learning. During the training process, the neural network model will learn how to predict the second audio signal formed at the target position based on the input first audio signal.

[0106] In a possible implementation, the second audio signal is determined based on the first audio signal, air-conditioning operating parameters when the microphone collects the first audio signal, and a pre-trained processing model.

[0107] Air conditioning operating parameters may include, but are not limited to, mode, wind speed, temperature setting, wind direction, and circulation. Modes may include cooling, heating, defrosting, and ventilation; wind speeds may include different gear information, such as 1, 2, 3, ..., and 7; temperature setting values ​​may include different set temperatures, such as 18°C, 19°C, ..., and 30°C; wind directions may include settings such as face, foot, and face + foot; and circulation may include internal and external circulation.

[0108] The first audio signal collected by the microphone and the air-conditioning operating parameters are input into a pre-trained neural network to obtain a second audio signal.

[0109] The pre-trained neural network is used to predict the second audio signal formed at the target position based on the first audio signal collected by the microphone under different air-conditioning operating parameters.

[0110] See also Figure 5 , a pre-trained neural network can be trained in the following way:

[0111] The system collects audio signals collected by a large number of microphones and the corresponding air-conditioning operating parameter data. At the same time, it collects the audio signals at the target position as labels and trains them through supervised learning. During the training process, the neural network model will learn how to predict the second audio signal formed at the target position based on the input first audio signal and air-conditioning operating parameters.

[0112] In other embodiments, see Figure 6 After the pre-trained neural network outputs the second audio signal formed at the target position, the first inverted audio signal is generated based on the second audio signal. This can be achieved by the modeling network. That is, the input of the modeling network is the first audio signal and the air conditioning operating parameters, and the output of the modeling network is the first inverted audio signal. Figure 7 The process of generating the second inverted audio signal can also be completed by a feedback adaptive filter. The feedback adaptive filter can achieve stable noise reduction. After the first audio signal is processed by low-pass filtering, it is input into the modeling network and the adaptive filter. The processed first audio signal passes through the modeling network to generate a first inverted audio signal, and the same signal passes through the feedback adaptive filter to generate a second inverted audio signal. The two are mixed to produce the actual speaker mixed output. The audio signal of the speaker mixed output will also be picked up by the microphone and will be used to update the feedback adaptive filter.

[0113] The modeling network can be, but is not limited to, an identification filter, a neural network, a linear model, a perceptron, a support vector machine, and the like.

[0114] It should be understood that the filters mentioned above are all digital filters. A digital filter is an algorithm or device composed of a digital multiplier, adder, and delay unit. The function of a digital filter is to process the digital code of the input discrete signal to achieve the purpose of changing the signal spectrum.

[0115] The modeling network, the preset transfer relationship, the pre-trained neural network, the first transfer function, and the second transfer function can be obtained by offline calibration.

[0116] The vehicle noise reduction method provided by the present invention, on the one hand, does not require additional sensors to be arranged in the air-conditioning duct, thereby significantly saving hardware costs, reducing system complexity, and avoiding difficulties in installing and assembling sensors; on the other hand, since it is non-duct noise control, there is no need to arrange speakers in the air-conditioning duct, thus avoiding difficulties in installing and assembling speakers; on the other hand, there is no need for additional hardware costs, the microphone can reuse the voice wake-up microphone in the cabin, and algorithms such as fixed modeling networks and adaptive filters can be executed in existing controllers such as power amplifiers or vehicle computers; on the fourth hand, a technical route combining fixed prediction and feedback adaptation is used. Due to the statistical stability of air-conditioning noise, fixed prediction can achieve relatively good results, and further combined with feedback adaptation can achieve more stable results.

[0117] Based on the same inventive concept, this embodiment also provides a vehicle noise reduction device, which can be applied to a controller on a vehicle or a vehicle noise reduction system on a vehicle. Figure 8 FIG. 6 is a block diagram of a vehicle noise reduction device according to an exemplary embodiment. The vehicle noise reduction device 600 may include:

[0118] A first processing module 601 is configured to obtain a first audio signal collected by a microphone in the vehicle, the first audio signal is used to determine a second audio signal of a target position in the vehicle, and the second audio signal is used to generate an inverted audio signal;

[0119] The second processing module 602 is configured to control the loudspeaker to play the inverted audio signal, so that the inverted audio signal can be used to reduce the second audio signal when it propagates to the target location.

[0120] Optionally, the inverted audio signal includes a first inverted audio signal obtained by inverting the second audio signal, and a second inverted audio signal determined according to the filtering parameter and the first audio signal.

[0121] Optionally, the filtering parameters are generated based on a second secondary acoustic feedback signal and a second noise audio signal, wherein the second noise audio signal represents an audio signal formed when the noise source propagates to the target position, and the second secondary acoustic feedback signal is an audio signal formed when the inverted audio signal played by the speaker is transmitted to the target position.

[0122] Optionally, the second noise audio signal is generated according to the first audio signal and a first secondary acoustic feedback signal, where the first secondary acoustic feedback signal is an audio signal formed when an inverted audio signal played by a loudspeaker is transmitted to the microphone.

[0123] Optionally, the vehicle noise reduction device 600 may further include:

[0124] A third processing module is configured to determine a first noise audio signal based on the first audio signal and the first secondary acoustic feedback signal, wherein the first noise audio signal represents an audio signal formed when the noise source propagates to the microphone;

[0125] The fourth processing module is configured to determine a second noise audio signal based on the first noise audio signal and a preset transfer relationship, wherein the preset transfer relationship represents a correspondence between the audio signal at the microphone and the audio signal at the target position.

[0126] Optionally, the first secondary acoustic feedback signal is determined based on an inverted audio signal played by the speaker and a pre-calibrated first transfer function, wherein the first transfer function characterizes acoustic path characteristics from the audio signal played by the speaker to the microphone; and / or,

[0127] The second secondary acoustic feedback signal is determined according to the inverted audio signal played by the loudspeaker and a pre-calibrated second transfer function, wherein the second transfer function represents the acoustic path characteristics of the audio signal played by the loudspeaker propagating to the target position.

[0128] Optionally, the first audio signal includes a plurality of sub-band signals in different frequency ranges, and the vehicle noise reduction device 600 may further include:

[0129] The fifth processing module is configured to determine, for each sub-band signal, a second inverse phase audio signal corresponding to the sub-band signal according to the filtering parameter and the sub-band signal.

[0130] Optionally, the second audio signal is determined according to the first audio signal and a pre-trained processing model.

[0131] Optionally, the second audio signal is determined according to the first audio signal, air-conditioning operating parameters when the microphone collects the first audio signal, and a pre-trained processing model.

[0132] Optionally, the target position is the ear position of a passenger in the vehicle.

[0133] Regarding the vehicle noise reduction device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the vehicle noise reduction method, and will not be elaborated on here.

[0134] The present disclosure also provides a controller, including:

[0135] a memory having a computer program stored thereon;

[0136] The processor is configured to execute the computer program in the memory to implement the steps of the above-mentioned vehicle noise reduction method.

[0137] The present disclosure also provides a vehicle noise reduction system, which includes:

[0138] microphone;

[0139] speaker;

[0140] The above-mentioned controller, microphone and speaker are all connected to the controller.

[0141] An embodiment of the present disclosure further provides a vehicle, comprising the above-mentioned controller or the above-mentioned vehicle noise reduction system.

[0142] The embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above-mentioned vehicle noise reduction method are implemented.

[0143] The embodiment of the present disclosure further provides a computer program product, including a computer program, which performs the steps of the above-mentioned vehicle noise reduction method when executed by a processor.

[0144] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings. However, the present disclosure is not limited to the specific details of the above embodiments. Within the technical concept of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all fall within the scope of protection of the present disclosure.

[0145] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0146] In addition, the various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A vehicle noise reduction method, characterized in that: The vehicle noise reduction method comprises: Acquire a first audio signal collected by a microphone in the vehicle, wherein the first audio signal is used to determine a second audio signal at a target position in the vehicle, and the second audio signal is used to generate an inverted audio signal; controlling a loudspeaker to play the inverted audio signal, so that the inverted audio signal can be used to reduce the second audio signal when propagating to the target location; The inverted audio signal includes a first inverted audio signal obtained by inverting the second audio signal, and a second inverted audio signal determined according to a filtering parameter and the first audio signal; The filtering parameters are generated based on a second secondary acoustic feedback signal and a second noise audio signal, wherein the second noise audio signal represents an audio signal formed when a noise source propagates to the target location, and the second secondary acoustic feedback signal is an audio signal formed when an inverted audio signal played by the speaker is transmitted to the target location; The filtering parameters are obtained as follows: The second noise audio signal is used as the desired signal and the error signal is used as the input signal. The error between the filter output and the desired signal is minimized by continuously adjusting the filter coefficients. The filter coefficients finally obtained by convergence are the desired filter parameters.

2. The vehicle noise reduction method according to claim 1, characterized in that: The second noise audio signal is generated according to the first audio signal and a first secondary acoustic feedback signal, where the first secondary acoustic feedback signal is an audio signal formed when the inverted audio signal played by the loudspeaker is transmitted to the microphone.

3. The vehicle noise reduction method according to claim 2, characterized in that: The vehicle noise reduction method further includes: Determine a first noise audio signal based on the first audio signal and the first secondary acoustic feedback signal, wherein the first noise audio signal represents an audio signal formed when a noise source propagates to the microphone; The second noise audio signal is determined based on the first noise audio signal and a preset transfer relationship, wherein the preset transfer relationship represents a correspondence between the audio signal at the microphone and the audio signal at the target position.

4. The vehicle noise reduction method according to claim 3, characterized in that: The first secondary acoustic feedback signal is determined based on the inverted audio signal played by the speaker and a pre-calibrated first transfer function, wherein the first transfer function represents the acoustic path characteristics of the audio signal played by the speaker to the microphone; and / or, The second secondary acoustic feedback signal is determined based on the inverted audio signal played by the loudspeaker and a pre-calibrated second transfer function, wherein the second transfer function represents the acoustic path characteristics of the audio signal played by the loudspeaker propagating to the target position.

5. The vehicle noise reduction method according to claim 1, characterized in that: The first audio signal includes a plurality of sub-band signals in different frequency ranges, and the vehicle noise reduction method further includes: For each of the sub-band signals, a second inverse phase audio signal corresponding to the sub-band signal is determined according to the filtering parameters and the sub-band signal.

6. The vehicle noise reduction method according to any one of claims 1 to 4, characterized in that: The second audio signal is determined according to the first audio signal and a pre-trained processing model.

7. The vehicle noise reduction method according to any one of claims 1 to 4, characterized in that: The second audio signal is determined according to the first audio signal, air-conditioning operating parameters when the microphone collects the first audio signal, and a pre-trained processing model.

8. The vehicle noise reduction method according to any one of claims 1 to 4, characterized in that: The target position is the ear position of the passenger in the car.

Images