Noise reduction processing method, headrest, seat and vehicle

Abstract

A noise reduction processing method, a headrest, a seat, and a vehicle, which relate to the technical field of noise reduction. The noise reduction processing method comprises: receiving at least one reference signal from a noise source acquisition apparatus and signals acquired by N+M error microphones, and selecting at least P undisturbed error microphones from the N+M error microphones; on the basis of signals acquired by the at least P error microphones and the at least one reference signal, generating a first noise suppression signal to cancel the noise, where both N and M are positive integers greater than 2, and P is a positive integer greater than 1 and less than or equal to N+M. Since the N+M error microphones are more widely distributed on headrests or seats, even if a certain error microphone is blocked by a human head or other objects or is disturbed by human behaviors, there are still at least P unblocked or undisturbed error microphones participating in the noise reduction operation, thereby improving the in-vehicle noise reduction effect.

Description

Noise reduction methods, headrests, seats and vehicles

[0001] This application claims priority to Chinese Patent Application No. 202411643228.2, filed on November 15, 2024, entitled "Noise Reduction Processing Method, Headrest, Seat and Vehicle", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of noise reduction technology, and in particular to a noise reduction processing method, headrest, seat and vehicle. Background Technology

[0003] As users' demands for in-car quietness gradually increase, active noise cancellation technology is increasingly being applied to various vehicle models across different price ranges. The principle of active noise cancellation is: noise is collected through a microphone, while a sound with the same amplitude but opposite phase is emitted by a speaker. The sound pressure levels of the two cancel each other out, thus achieving noise reduction.

[0004] However, in practical applications, factors such as microphone obstruction and interference noise near the microphone can lead to a decrease in the active noise cancellation effect of the vehicle, or even an increase in noise instead of reduction, affecting the passenger experience. Summary of the Invention

[0005] This application provides a noise reduction method, a headrest, a seat, and a vehicle, which can improve the noise reduction effect of a vehicle.

[0006] In a first aspect, embodiments of this application provide a noise reduction processing method, the method comprising: receiving at least one reference signal and signals collected from N+M error microphones, where N and M are both positive integers greater than 2; generating a first noise suppression signal based on signals collected from at least P error microphones among the N+M error microphones and at least one reference signal; using the signals from at least P error microphones for noise reduction processing, where P is a positive integer greater than 1 and less than or equal to N+M; and outputting the first noise suppression signal.

[0007] In this embodiment, the N+M error microphones can be microphones located on the front of the headrest or seat. These error microphones can also be called real microphones, used to collect noise signals near the human ear. The signals collected by the error microphones participating in the noise reduction processing are also called error signals. The reference signal is a signal near the noise source, obtained from a noise source acquisition device, which may include, for example, a microphone, an accelerometer, or a speed sensor.

[0008] In the noise reduction processing scheme shown in this embodiment, since the N+M error microphones are distributed over a wider range on the headrest or seat, even if a certain error microphone is blocked by a person's head (or other things, such as clothing, cushions, etc.) or interfered with by a person's behavior (such as patting, rubbing, etc.), there are still at least P error microphones that are not blocked or interfered with. Therefore, the first noise suppression signal generated based on the signals collected by at least P error microphones and at least one reference signal has a better noise reduction effect.

[0009] In an alternative embodiment of the first aspect, the method further includes: determining at least P undisturbed error microphones from the N+M error microphones based on at least one of the signals acquired by the N+M error microphones and the signals acquired by the sensor.

[0010] Among the interferences, the error microphone is affected by: wind, being blocked by a person's head, voice interference, friction noise, and slapping noise.

[0011] Optionally, the sensor includes at least one of an image sensor, an infrared sensor, and an ultrasonic sensor.

[0012] In some embodiments, based on the signals collected by N+M error microphones, frequency domain analysis is performed on the signals collected by N+M error microphones to determine at least P error microphones that have not been interfered with.

[0013] In some embodiments, based on the signals acquired by N+M error microphones, at least P undisturbed error microphones are determined from the N+M error microphones by calculating the correlation between the signals acquired by each pair of error microphones. Here, the correlation refers to the degree of similarity between two signals.

[0014] In an optional embodiment of the first aspect, determining at least P uninterrupted error microphones from the N+M error microphones based on signals collected by N+M error microphones includes: removing target microphones that satisfy a first condition from the N+M error microphones to determine at least P error microphones; the first condition includes at least one of the following: the decibel value corresponding to the signal attenuation in a first frequency band is greater than a first threshold; the energy value of the signal collected by the target microphone in the full frequency band or a second frequency band is greater than a second threshold; the correlation between the signal collected by the target microphone and the signals collected by other error microphones is less than a third threshold.

[0015] For example, the first frequency band can be 500 to 20000 Hz, and the second frequency band can be 1000 to 8000 Hz.

[0016] If the decibel value corresponding to the attenuation in the first frequency band of the signal collected by a certain error microphone is greater than the first threshold, it indicates that there is attenuation in the mid-to-high frequency range of the signal from the error microphone, and it can be considered that the error microphone is blocked, such as by a person's head.

[0017] If the energy value of the signal collected by a certain error microphone in the full frequency band or the second frequency band is greater than the second threshold, it indicates that there is a friction sound signal in the signal collected by the error microphone, and it can be considered that the error microphone is interfered with by friction sound, such as the friction of clothing.

[0018] If the correlation between the signal collected by a certain error microphone and the signals collected by other error microphones on the headrest or seat is less than the third threshold, it indicates that the error microphone is being interfered with.

[0019] In summary, the first condition may include multiple conditions such as the error microphone being interfered with, and this application embodiment does not limit this.

[0020] If a certain error microphone meets one of the first conditions, it can be determined that the error microphone is interfered with and can be removed. The signal collected by the error microphone will not participate in the subsequent noise reduction processing operation, so as to improve the accuracy of the noise reduction processing operation and thus improve the noise reduction effect.

[0021] In one alternative embodiment of the first aspect, determining at least P undisturbed error microphones from N+M error microphones based on signals acquired by sensors includes: determining a first relative positional relationship between a person's head and a headrest or seat based on signals acquired by sensors; and removing target microphones obscured by the person's head from the N+M error microphones based on the first relative positional relationship and the positional information of the N+M error microphones on the headrest or seat, thereby determining at least P error microphones.

[0022] In this embodiment, the first relative positional relationship between the head and the headrest or seat can refer to the positional information of the headrest area obscured by the head and the positional information of the headrest area not obscured by the head. If the positional information of a certain error microphone on the headrest overlaps with the positional information of the headrest area obscured by the head, it indicates that the error microphone is obscured by the head.

[0023] In some embodiments, the first relative positional relationship between the head and the headrest or seat may refer to the distance between the back of the head and the front of the headrest in a first direction and the position information of the back of the head in the headrest area. The first direction is perpendicular to the plane of the front of the headrest. If the position information of an error microphone on the headrest overlaps with the position information of the back of the head in the headrest area, and the distance between the back of the head and the target microphone in the first direction is less than a certain distance threshold, it indicates that the error microphone is obstructed by the head.

[0024] By analyzing the signals collected by the sensors, the relative positional relationship between the person's head and N+M error microphones located on the front of the headrest or seat can be determined. Based on this relative positional relationship, error microphones obscured by the person's head are eliminated. This prevents signals from interference-laden error microphones from participating in subsequent noise reduction processing, thereby improving the accuracy of the noise reduction process and ultimately enhancing the noise reduction effect.

[0025] In one optional embodiment of the first aspect, determining at least P undisturbed error microphones from N+M error microphones based on signals collected by sensors includes: determining a second relative positional relationship between the human ear and a headrest or seat based on signals collected by sensors; and, based on the second relative positional relationship and the positional information of the N+M error microphones on the headrest or seat, removing target microphones from the N+M error microphones whose distance from the human ear is greater than or equal to a preset distance value, so as to determine at least P error microphones.

[0026] In this embodiment, the second relative positional relationship between the human ear and the headrest or seat can refer to the positional information of the human ear projected onto the headrest area. Based on the positional information of the human ear projected onto the headrest area and the positional information of a certain error microphone on the headrest, the distance value between the error microphone and the human ear on the headrest plane can be determined; if the distance value is greater than or equal to a preset distance value, it indicates that the error microphone is far from the human ear; the error microphone can be discarded, that is, the signal collected by the error microphone will not participate in the subsequent noise reduction processing operation.

[0027] By analyzing the signals collected by the sensors, the relative positional relationship between the human ear and N+M error microphones placed on the front of the headrest or seat can be determined. Based on this relative positional relationship, error microphones that are far away from the human ear can be eliminated, and the signals collected by error microphones that are closer to the human ear can be used for noise reduction calculation, which can improve the noise reduction effect.

[0028] In one optional embodiment of the first aspect, the N+M error microphones include: a first microphone, a second microphone, a third microphone, and a fourth microphone; based on a second relative positional relationship and the positional information of the N+M error microphones on the headrest or seat, target microphones whose distance value from the human ear is greater than a preset distance value are removed from the N+M error microphones, including: if the first distance value between the first microphone and the left ear is less than the preset distance value, the second distance value between the second microphone and the right ear is less than the preset distance value, the third distance value between the third microphone and the left ear is greater than or equal to the preset distance value, and the fourth distance value between the fourth microphone and the right ear is greater than or equal to the preset distance value, then the third microphone and the fourth microphone are removed.

[0029] Referring to Figure 13, the first microphone corresponds to error microphone 1, the second microphone corresponds to error microphone 2, the third microphone corresponds to error microphone 3, and the fourth microphone corresponds to error microphone 4. As shown in Figure 13(a), since error microphones 3 and 4 are far from the human ear, they can be discarded, meaning that the signals collected by error microphones 3 and 4 do not participate in subsequent noise reduction processing.

[0030] By analyzing the signals collected by the sensors, the relative positional relationship between the human ear and the four error microphones set in front of the headrest or seat can be determined. Based on this relative positional relationship, error microphones that are far away from the human ear can be eliminated, and the signals collected by error microphones that are close to the human ear can be used for noise reduction calculation to achieve adaptive single-area (the upper half of the headrest area as shown in Figure 13(a)) noise reduction, which can improve the noise reduction effect.

[0031] In one optional embodiment of the first aspect, a first noise suppression signal is generated based on signals collected by at least P error microphones out of N+M error microphones and at least one reference signal, including: determining a second relative positional relationship between a human ear and a headrest or seat based on signals collected by sensors; determining weight values ​​corresponding to the signals collected by at least P error microphones based on the second relative positional relationship and position information of at least P error microphones on the headrest or seat; and generating the first noise suppression signal based on the weight values ​​corresponding to the signals collected by at least P error microphones, the signals collected by at least P error microphones, and at least one reference signal.

[0032] In this embodiment, based on the second relative positional relationship between the human ear and the headrest or seat, and the positional information of each error microphone on the headrest or seat, the distance between each error microphone and the human ear on the headrest plane can be determined. Then, the weight value corresponding to the signal collected by each error microphone can be adjusted based on this distance value. The closer the distance, the closer the signal collected by the corresponding error microphone is to the noise signal near the human ear, thus increasing the weight value of that signal; the farther the distance, the greater the deviation between the signal collected by the corresponding error microphone and the noise signal near the human ear, thus decreasing the weight value of that signal.

[0033] In some embodiments, filter parameters can be updated using the following formula:

[0034] In the formula, This represents the filter parameters at time n+1. Let Ω represent the filter parameters at time n, Ω represent the set of uninterrupted error microphones, and α represent the filter parameters at time n. m e represents the weight value corresponding to the signal (error signal) collected by the m-th error microphone in the set.m (n) represents the signal acquired by the m-th error microphone at time n, where m ranges from 1 to P. μ is the update factor, which takes values ​​from (0,1]. The larger the value, the larger the filter parameter update. This is the filtered reference signal.

[0035] By analyzing the signals collected by the sensors, the relative positional relationship between the human ear and at least P undisturbed error microphones placed on the front of the headrest or seat can be determined. Then, based on the relative positional relationship, the weight values ​​corresponding to the signals collected by the at least P error microphones can be adjusted to optimize the noise reduction processing algorithm and improve the noise reduction effect.

[0036] In one alternative embodiment of the first aspect, generating a first noise suppression signal based on signals acquired by at least P error microphones out of N+M error microphones and at least one reference signal includes: generating a signal from a virtual microphone near the human ear based on signals acquired by at least P error microphones; and generating the first noise suppression signal based on the signal from the virtual microphone and at least one reference signal.

[0037] In this embodiment, the signals from the virtual microphones near the ears include signals from the virtual microphones near the left ear and the virtual microphones near the right ear. The signals from the virtual microphones near the ears can be estimated from signals collected by at least P interference-free error microphones on the headrest or seat.

[0038] In some embodiments, after determining the signal from the virtual microphone near the ear, the filter parameters can be updated using the following formula:

[0039] In the formula, This represents the filter parameters at time n+1. Let e1 represent the filter parameters at time n, where μ is the update factor. ′ (n) represents the signal from the virtual microphone near the left ear, α1 represents the weight value of the signal from the virtual microphone near the left ear, and e2 represents the weight value of the signal from the virtual microphone near the left ear. ′ (n) represents the signal from the virtual microphone near the right ear, and α2 represents the weight value of the signal from the virtual microphone near the right ear. This is the filtered reference signal.

[0040] Referring to Figure 15, after determining the updated filter parameters based on Formula 2, the filter parameters are updated by controlling the filter output. The control signal at time n is obtained from the filter output, amplified by the power amplifier, and then the first noise suppression signal is obtained. The first noise suppression signal is then output through the speaker.

[0041] The noise reduction processing scheme illustrated in the above embodiments utilizes signals collected by multiple interference-free error microphones on the seat or headrest to estimate the signal of a virtual microphone near the ear. Based on the signal from the virtual microphone near the ear and at least one reference signal, the filter parameters are updated. Then, based on the updated filter parameters, a first noise suppression signal is generated. Since the error microphones participating in signal estimation are all interference-free and there are multiple error microphones, the signal estimation accuracy is higher and the robustness is better, providing data support for subsequent noise reduction calculations and improving the noise reduction effect.

[0042] In one optional embodiment of the first aspect, generating a signal of a virtual microphone near a human ear based on signals collected by at least P error microphones includes: inputting the signals collected by at least P error microphones into a pre-trained first model corresponding to at least P error microphones, and processing the signals by the first model to obtain a signal of a virtual microphone near the left ear; and inputting the signals collected by at least P error microphones into a pre-trained second model corresponding to at least P error microphones, and processing the signals by the second model to obtain a signal of a virtual microphone near the right ear.

[0043] In this embodiment, the first model is used to estimate the signal of the virtual microphone near the left ear, and the second model is used to estimate the signal of the virtual microphone near the right ear. Both the first and second models are trained based on a large amount of training data.

[0044] The first model may include multiple models, such as Model 1 and Model 2. For example, the input to Model 1 includes signals collected by real microphones 1 to 3 in Figure 16a or Figure 16b, and Model 1 estimates the signal of the virtual microphone 5 near the left ear based on the signals collected by real microphones 1 to 3. The input to Model 2 includes signals collected by real microphones 1, 2, and 4 in Figure 16a or Figure 16b, and Model 2 estimates the signal of the virtual microphone 5 near the left ear based on the signals collected by real microphones 1, 2, and 4.

[0045] The second model can include multiple models, such as Model 3, Model 4, etc. For example, the input to Model 3 includes signals collected by real microphones 1 to 3 in Figure 16a or Figure 16b. Model 3 estimates the signal of a virtual microphone (not shown) near the right ear based on the signals collected by real microphones 1 to 3. The input to Model 4 includes signals collected by real microphones 2, 3, and 4 in Figure 16a or Figure 16b. Model 4 estimates the signal of a virtual microphone near the right ear based on the signals collected by real microphones 2, 3, and 4.

[0046] In an optional embodiment of the first aspect, receiving at least one reference signal and signals collected from N+M error microphones includes: at a first moment, receiving at least one reference signal and signals collected from N+M error microphones; the method further includes: at a second moment, if a second condition is met, maintaining the output of a first noise suppression signal; the second moment is later than the first moment, and the second condition includes at least one of the following: detecting that at least one error microphone is being blown by wind, or detecting that the signal from at least one error microphone contains a human voice signal.

[0047] The noise reduction processing scheme shown in this embodiment can maintain the noise suppression signal of the previous moment without updating the filter parameters when wind noise or voice interference is detected, thus avoiding the decrease in noise reduction effect caused by wind noise or voice interference.

[0048] Secondly, embodiments of this application provide a headrest, including: M+N error microphones, M error microphones located in the left front region of the headrest, N error microphones located in the right front region of the headrest, and the pickup holes of the M+N error microphones facing the front of the headrest; N and M are both positive integers greater than 2.

[0049] Multiple error microphones are set in the left and right front areas of the headrest. Compared with the conventional microphone setting (one error microphone near each ear), the head is less likely to block all error microphones on one side (such as the left or right front side of the headrest) at the same time, which can achieve better noise reduction effect.

[0050] In an alternative embodiment of the second aspect, the headrest includes: a first microphone and a second microphone, the first microphone being located in the left front region of the headrest and the second microphone being located in the right front region of the headrest, the pickup hole of the first microphone and the pickup hole of the second microphone being set at a horizontal distance of 15cm to 30cm.

[0051] It should be noted that the microphone's pickup hole is typically located at the microphone's geometric center. In some embodiments, the horizontal distance between the pickup hole of the first microphone and the pickup hole of the second microphone is set between 15cm and 30cm, which can also be described as: the horizontal distance between the geometric center of the first microphone and the geometric center of the second microphone is set between 15cm and 30cm.

[0052] By constraining the horizontal distance between the error microphones in the left and right areas of the front of the headrest, the error microphones in the left and right areas are made to be as close as possible to the human ear, while also having a certain span, so as to avoid being blocked by the human head (usually the back of the head) during normal use.

[0053] In an optional embodiment of the second aspect, the headrest includes a third microphone and a fourth microphone, both of which are located in the left front region or the right front region of the headrest, and the third microphone and the fourth microphone are two adjacent microphones, with the distance between the pickup hole of the third microphone and the pickup hole of the fourth microphone set between 5cm and 15cm.

[0054] By constraining the distance between two adjacent error microphones in the left (or right) front area of ​​the headrest, the error microphones are positioned as close to the ear as possible while maintaining a certain distance to prevent simultaneous blockage of error microphones in either the left or right front area. This ensures that regardless of head movement, at least one error microphone in either the left or right front area remains unobstructed.

[0055] In an alternative embodiment of the second aspect, the headrest further includes: a speaker located on the front side of the headrest and facing forward of the headrest, wherein the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone is greater than or equal to 3 cm.

[0056] Noise reduction can be achieved by placing speakers facing forward on the front of the headrest. These speakers output noise suppression signals to cancel out in-vehicle noise. The distance between the speakers and each error microphone located on the front of the headrest should be greater than a certain distance (e.g., 3cm) to avoid problems such as excessively large error microphone signal amplitude caused by speaker approach instability.

[0057] Thirdly, embodiments of this application provide a seat, including a headrest as described in the second aspect. The third aspect of this application corresponds to the technical solution of the second aspect, and the beneficial effects achieved are similar, so further details are omitted.

[0058] Fourthly, embodiments of this application provide a seat, including: M+N error microphones, M error microphones located on the upper part of the left front region of the seat, N error microphones located on the upper part of the right front region of the seat, and the pickup holes of the M+N error microphones facing the front of the seat; N and M are both positive integers greater than 2.

[0059] Multiple error microphones are set in the upper left and upper right front areas of the seat. Compared with the conventional microphone setting (one error microphone in each of the upper left and upper right front areas of the seat), the head is less likely to block all error microphones on one side (such as the left or right front side of the headrest) at the same time, which can achieve better noise reduction effect.

[0060] In an alternative embodiment of the fourth aspect, the seat includes a fifth microphone and a sixth microphone, the fifth microphone being located in the upper part of the left front region of the seat and the sixth microphone being located in the upper part of the right front region of the seat, the pickup hole of the fifth microphone and the pickup hole of the sixth microphone being set at a horizontal distance of 15cm to 30cm.

[0061] In some embodiments, the horizontal distance between the pickup hole of the fifth microphone and the pickup hole of the sixth microphone is set between 15cm and 30cm, or it can be described as: the horizontal distance between the geometric center of the fifth microphone and the geometric center of the sixth microphone is set between 15cm and 30cm.

[0062] By constraining the horizontal distance between the error microphones in the upper front and left areas of the seat, the error microphones in the left and right areas are made to be as close as possible to the human ear, while also having a certain span, so as to avoid being blocked by the human head (usually the back of the head) during normal use.

[0063] In an alternative embodiment of the fourth aspect, the seat includes a seventh microphone and an eighth microphone, both of which are located in the upper part of the right front area or the left front area of ​​the seat, and the seventh microphone and the eighth microphone are two adjacent microphones, with the distance between the pickup hole of the seventh microphone and the pickup hole of the eighth microphone set between 5cm and 15cm.

[0064] By constraining the distance between two adjacent error microphones in the upper left (or right) front area of ​​the seat, the error microphones are positioned as close to the listener's ear as possible while maintaining a certain distance to prevent simultaneous blockage of error microphones in either the left or right front area. This ensures that regardless of head movement, at least one error microphone in either the left or right front area remains unobstructed.

[0065] In an alternative embodiment of the fourth aspect, the seat further includes: a speaker located on the upper front side of the seat and facing forward, wherein the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone is greater than or equal to 3 cm.

[0066] Noise reduction can be achieved by installing speakers facing the headrest on the upper front side of the seat. These speakers output noise suppression signals to cancel out interior noise. The distance between the speakers and each error microphone located on the front side of the seat should be greater than a certain distance (e.g., 3cm) to avoid problems such as excessively large error microphone signal amplitude caused by speaker approach instability.

[0067] Fifthly, embodiments of this application provide a noise reduction processing apparatus, including: a processor and a memory; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory, causing the noise reduction processing apparatus to perform the method as described in the first aspect.

[0068] Sixthly, embodiments of this application provide a vehicle, including: a seat as described in the third or fourth aspect and a noise reduction processing device as described in the fifth aspect.

[0069] In a seventh aspect, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as described in the first aspect.

[0070] Eighthly, embodiments of this application provide a chip system including at least one processor and a communication interface, the communication interface and at least one processor being interconnected via a line, the at least one processor being used to run computer programs or instructions to perform the method as described in the first aspect.

[0071] Ninthly, embodiments of this application provide a computer program product, including a computer program that, when run, causes a computer to perform the method as described in the first aspect of the claim.

[0072] The fifth to ninth aspects of this application correspond to the technical solutions of the first aspect of this application and achieve similar beneficial effects, which will not be repeated here. Attached Figure Description

[0073] Figure 1 is a schematic diagram of the principle of active noise reduction provided in the embodiment of this application;

[0074] Figure 2 is a schematic diagram of one location of the error microphone provided in an embodiment of this application inside a vehicle;

[0075] Figure 3a is a schematic diagram of one position of the error microphone provided in an embodiment of this application in a vehicle seat;

[0076] Figure 3b is a schematic diagram of another position of the error microphone provided in the embodiment of this application in the vehicle seat;

[0077] Figure 4 is a schematic diagram of the vehicle-mounted active noise cancellation system provided in an embodiment of this application;

[0078] Figure 5 is a schematic diagram of an error microphone installed on a headrest according to an embodiment of this application;

[0079] Figure 6a is a schematic diagram of the position of the human ear and the error microphone provided in an embodiment of this application;

[0080] Figure 6b is a schematic diagram of another position of the human ear and the error microphone provided in an embodiment of this application;

[0081] Figure 7a is a schematic diagram showing the position of the error microphone in the headrest front view according to an embodiment of this application;

[0082] Figure 7b is a schematic diagram showing the position of the error microphone provided in the embodiment of this application in the left view of the headrest;

[0083] Figure 8 is another schematic diagram of an error microphone installed on a headrest according to an embodiment of this application;

[0084] Figure 9 is another schematic diagram of an error microphone installed on a headrest according to an embodiment of this application;

[0085] Figure 10 is a schematic diagram of the constraint conditions between error microphones and between error microphones and speakers provided in the embodiments of this application;

[0086] Figure 11 is a flowchart of a noise reduction processing method provided in an embodiment of this application;

[0087] Figure 12 is a schematic diagram of a noise reduction processing algorithm provided in an embodiment of this application;

[0088] Figure 13a is a schematic diagram of a setting of the weight value corresponding to the signal collected by the error microphone according to an embodiment of this application;

[0089] Figure 13b is a schematic diagram of another setting of the weight value corresponding to the signal collected by the error microphone provided in the embodiment of this application;

[0090] Figure 14 is a flowchart of another noise reduction processing method provided in an embodiment of this application;

[0091] Figure 15 is a schematic diagram of another noise reduction processing algorithm provided in an embodiment of this application;

[0092] Figure 16a is a schematic diagram of an embodiment of this application for estimating virtual microphone signals near the left ear;

[0093] Figure 16b is another schematic diagram provided by an embodiment of this application for estimating virtual microphone signals near the left ear;

[0094] Figure 17 is a flowchart of another noise reduction processing method provided in an embodiment of this application;

[0095] Figure 18 is a schematic diagram of a noise reduction processing device provided in an embodiment of this application;

[0096] Figure 19 is a schematic diagram of the structure of a chip provided in an embodiment of this application. Detailed Implementation

[0097] With the rapid development of vehicle technology, people's demands for ride comfort have increased, making vehicle noise reduction performance an important indicator for users when choosing a vehicle. In-vehicle noise control includes passive and active control. Passive control mainly reduces noise generation and propagation through physical means such as modifying the vehicle's interior structure, adding damping materials, or using shock absorbers and vibration dampers. Active control, also known as active noise control, cancels noise by introducing inverse sound waves.

[0098] Taking active noise cancellation for road noise as an example, Figure 1 shows a schematic diagram of the principle of active noise cancellation. As shown in Figure 1, the vehicle uses a microphone to collect road noise signals and emits sound waves with waveforms opposite to these road noise signals through a speaker. The two cancel each other out, thereby significantly reducing the noise inside the vehicle and achieving active noise cancellation. Active noise cancellation technology has advantages such as significant control effect on mid-to-low frequency noise, lightweight system, and strong real-time performance.

[0099] The microphone setup has a significant impact on noise reduction effectiveness. Generally, the closer the microphone is to the ear, the wider the noise reduction bandwidth and the greater the total noise reduction.

[0100] In some embodiments, the microphone can be placed on the vehicle body near the person's head, such as on the roof, or on the A / B / C / D pillars of the vehicle, as shown in Figure 2.

[0101] In some embodiments, the microphone can be placed on the seat near the person's head, such as on the headrest or shoulder. As shown in Figure 3a, the microphone can be placed on the headrest of a seat with an independent headrest, or near the shoulder of the seat; as shown in Figure 3b, the microphone can be placed near the head of an integrated seat. Because the microphone placed on the seat is closer to the ear, and the distance between the microphone and the ear does not change much when the seat moves forward, backward, up, down, or the backrest angle changes, it has a better noise reduction effect.

[0102] In this embodiment of the application, the microphone placed on the seat or headrest near the person's head is referred to as an error microphone.

[0103] Based on this, this application embodiment shows an in-vehicle active noise cancellation system, as shown in Figure 4. The system includes: a noise source acquisition device disposed near the noise source, an error microphone disposed on the seat or headrest, a noise cancellation processing device, a power amplifier, and a speaker.

[0104] Noise source acquisition devices are used to collect signals from the vicinity of the noise source (such as the engine location, air conditioning vents, or outside the vehicle). These signals are generally referred to as reference signals. For example, a noise source acquisition device could be a microphone in active noise-canceling headphones, an accelerometer in a road noise active noise cancellation system, or a speed sensor in an engine noise active noise cancellation system. Feedforward-type active noise cancellation systems require a reference signal for noise reduction calculations; feedback-type active noise cancellation systems do not require a reference signal. Currently, the vast majority of in-vehicle active noise cancellation systems are feedforward-type active noise cancellation systems.

[0105] An error microphone is used to collect noise signals near the ear; this signal is generally called the error signal, and it is the signal that needs to be noise-reduced. Active noise cancellation systems also require the error signal to participate in the noise reduction calculation.

[0106] In some embodiments, the error microphone may also be referred to as a noise-canceling microphone.

[0107] The noise reduction processing unit is the core of the active noise cancellation system. It receives reference signals from the noise source acquisition device and error signals from the error microphone, and processes these signals using a preset active noise cancellation algorithm to generate control signals for noise cancellation in real time. The active noise cancellation algorithm will be discussed later and will not be elaborated here.

[0108] In some embodiments, the noise reduction processing device may also be referred to as an active noise reduction controller.

[0109] After the control signal is amplified by the power amplifier, it is used by the loudspeaker to generate a control sound field, which is superimposed on the original noise field to achieve a noise reduction effect in the target area near the ear.

[0110] Speakers can be installed on the vehicle body and in the headrests or shoulders of the seats.

[0111] Based on the aforementioned active noise cancellation system, Figure 5 shows a schematic diagram of an error microphone mounted on the headrest. Figure 5 uses a standalone headrest as an example, with one error microphone positioned in the left front area and one in the right front area of ​​the headrest to collect noise signals near the left and right ears, respectively. In some embodiments, one error microphone can also be positioned in the upper left front area and the upper right front area of ​​the seat to collect noise signals near the left and right ears, respectively.

[0112] However, error microphones mounted on seats or headrests also have certain limitations. Because the error microphones are fixed to the seat or headrest and are close to the user's head, the following problems can easily occur:

[0113] 1) Error microphone is blocked. Due to movement of the head and shoulders, as well as clothing and pillows, there is a high probability that the error microphone will be blocked, leading to a decrease in noise reduction effect, or even an increase in noise instead of reduction. For example, as shown in Figure 5, if the head moves a certain distance to the left, the head will block the error microphone on the left side of the headrest.

[0114] 2) Distance between the ear and the error microphone. The noise reduction effect at the ear decreases as the distance between the ear and the error microphone increases. Due to factors such as the person's height, sitting posture, or headrest adjustment, the ear will deviate from the optimal noise reduction position, resulting in a decrease in noise reduction effectiveness. For example, in Figure 6a, the ear is at the same height as the error microphones in the left and right front areas of the headrest, indicating that the ear is in the optimal noise reduction position. In Figure 6b, the ear is a certain distance above the error microphones in the left and right front areas of the headrest, indicating that the ear has deviated from the optimal noise reduction position.

[0115] 3) Error microphone interference noise. People's heads, clothing, hands, and other parts are likely to rub, touch, or knock on the error microphone, thereby generating interference noise signals at the error microphone (humans do not perceive this interference noise obviously, but the error microphone will have a large interference response). If the vehicle's active noise cancellation system reduces this noise, similar loud interference signals will be generated by the speakers, affecting the passenger experience.

[0116] Based on the above analysis, if only a set of error microphones are symmetrically placed on the seat or headrest, as shown in Figure 5, if one of the error microphones is blocked by a person's head (or other objects such as clothing, cushions, etc.) or interfered with by human actions (such as patting, rubbing, etc.), the noise reduction effect will be worse, and may even increase the noise problem. In addition, if the passenger's head position is too high or too low due to their height or sitting posture, the noise reduction effect will also be affected because the person's ears are far away from the error microphones.

[0117] Therefore, this application embodiment illustrates a method for setting up error microphones, in which an error microphone array is set up on the front side of the vehicle seat or headrest, such as setting at least two error microphones in the right front area of ​​the seat or headrest and at least two error microphones in the left front area. Compared to setting one error microphone each in the right front area and left front area of ​​the seat or headrest, since the error microphones cover a wider area, in most scenarios it is impossible to simultaneously block all error microphones on the same side (such as the left or right side of the headrest). Combined with the noise reduction processing algorithm proposed in this application after blocking the error microphones (see below for details), a better noise reduction effect can be maintained.

[0118] Compared to the error microphone setup shown in Figure 5, the presence of multiple error microphones on the front of the seat or headrest increases the coverage area, enabling signal acquisition over a wider range and improving noise reduction.

[0119] The following detailed description of the error microphone array setup in this case will be provided using several specific embodiments.

[0120] (a) The error microphone array is set in front of the headrest.

[0121] This application provides a headrest comprising M+N error microphones. M error microphones are located in the left front region of the headrest, and N error microphones are located in the right front region of the headrest. The pickup holes of the M+N error microphones face forward of the headrest to collect noise signals near the left and right ears, respectively. Here, N and M are both positive integers greater than 2.

[0122] In one example of this embodiment, Figure 7a shows a schematic diagram of the position of the error microphones in the front view of the headrest, and Figure 7b shows a schematic diagram of the position of the error microphones in the left view of the headrest. Figures 7a and 7b use an independent headrest as an example. Two error microphones are arranged in the left front area of ​​the headrest, as error microphones 1 and 3 in Figure 7a or 7b, and two error microphones are arranged in the right front area of ​​the headrest, as error microphones 2 and 4 in Figure 7a or 7b. That is, a total of four error microphones are arranged on the front side of the headrest, and the pickup holes of all four error microphones face the front of the headrest. The four error microphones are stationary relative to the headrest but can move with the headrest.

[0123] In one example of this embodiment, Figure 8 shows another schematic diagram of the error microphones mounted on the headrest. Figure 8 uses a standalone headrest as an example, with two error microphones (error microphones 1 and 3) positioned on the left front side of the headrest, and one error microphone (error microphone 2) positioned on the right front side. That is, a total of three error microphones are mounted on the front side of the headrest, with the pickup holes of all three microphones facing forward. The three error microphones remain stationary relative to the headrest but can move with it.

[0124] In one example of this embodiment, Figure 9 shows another schematic diagram of the error microphones disposed on the headrest. Figure 9 uses a standalone headrest as an example, with three error microphones (1, 3, and 5) disposed on the left front side of the headrest, and three error microphones (2, 4, and 6) disposed on the right front side of the headrest. That is, a total of six error microphones are disposed on the front side of the headrest, with the pickup holes of all six microphones facing forward. The six error microphones remain stationary relative to the headrest but can move with it.

[0125] In some embodiments, the headrest includes a first microphone and a second microphone. The first microphone is located in the left front region of the headrest, and the second microphone is located in the right front region of the headrest. The horizontal distance between the pickup holes of the first microphone and the second microphone is set between 15cm and 30cm. That is, the maximum horizontal distance between the pickup holes of the first microphone and the second microphone can be set to 15cm, and the minimum horizontal distance between the pickup holes of the first microphone and the second microphone can be set to 30cm.

[0126] It should be noted that the microphone's pickup hole is typically located at the microphone's geometric center. In some embodiments, the horizontal distance between the pickup hole of the first microphone and the pickup hole of the second microphone is set between 15cm and 30cm, which can also be described as: the horizontal distance between the geometric center of the first microphone and the geometric center of the second microphone is set between 15cm and 30cm.

[0127] For example, in Figure 10, error microphone 1 and error microphone 2 are at the same height on the headrest. The horizontal distance between the pickup holes of error microphone 1 and error microphone 2 is 'a', which can be set between 15cm and 30cm, for example, 22cm. In Figure 10, error microphone 3 and error microphone 4 are at the same height on the headrest. The horizontal distance between the pickup holes of error microphone 3 and error microphone 4 can be set between 15cm and 30cm, for example, 22cm.

[0128] Similarly, in Figure 8, error microphone 1 and error microphone 2 are at the same height on the headrest, and the horizontal distance between the pickup holes of error microphone 1 and error microphone 2 is set between 15 and 30 cm.

[0129] Similarly, in Figure 9, error microphones 1 and 2 are at the same height on the headrest, as are error microphones 3 and 4, and error microphones 5 and 6. The horizontal distance between the pickup holes of error microphones 1 and 2 is a1, a2, and a3 (not shown in the figure) is a2 between the pickup holes of error microphones 3 and 4. All three distances can be set between 15 and 30 cm.

[0130] The distance between human ears is usually around 16cm. By constraining the horizontal distance of the error microphones in the left and right areas of the front of the headrest, the error microphones in the left and right areas are made to be as close as possible to the human ear, while also having a certain span, so as to avoid being blocked by the human head (usually the back of the head) during normal use.

[0131] In some embodiments, the headrest includes a third microphone and a fourth microphone, both of which are located in the left front region or the right front region of the headrest, and the third microphone and the fourth microphone are two adjacent microphones, with the distance between the pickup hole of the third microphone and the pickup hole of the fourth microphone set between 5cm and 15cm.

[0132] For example, continuing to refer to Figure 10, error microphone 1 and error microphone 3 are two adjacent microphones located in the left front area of ​​the headrest. The distance between the pickup hole of error microphone 1 and the pickup hole of error microphone 3 is b, which can be set between 5cm and 15cm, for example, 10cm. In Figure 10, error microphone 2 and error microphone 4 are two adjacent microphones located in the right front area of ​​the headrest. The distance between the pickup hole of error microphone 2 and the pickup hole of error microphone 4 can be set between 5cm and 15cm, for example, 10cm.

[0133] Similarly, in Figure 8 or Figure 9, error microphone 1 and error microphone 3 are two adjacent microphones located in the left front area of ​​the headrest, with the distance between the pickup holes of error microphone 1 and error microphone 3 set between 5cm and 15cm. In Figure 9, error microphone 3 and error microphone 5 are two adjacent microphones located in the left front area of ​​the headrest, with the distance between the pickup holes of error microphone 3 and error microphone 5 set between 5cm and 15cm.

[0134] In Figure 9, error microphones 2 and 4 are two adjacent microphones located on the right front side of the headrest, with a distance b1 between the pickup holes of error microphone 2 and error microphone 4. Similarly, error microphones 4 and 6 are also located on the right front side of the headrest, with a distance b2 between the pickup holes of error microphone 4 and error microphone 6. Both b1 and b2 can be set between 5cm and 15cm.

[0135] By constraining the distance between two adjacent error microphones in the left (or right) front area of ​​the headrest, the error microphones are positioned as close to the ear as possible while maintaining a certain distance to prevent simultaneous blockage of error microphones in either the left or right front area. This ensures that regardless of head movement, at least one error microphone in either the left or right front area remains unobstructed.

[0136] In some embodiments, the headrest further includes a speaker. The speaker is located on the front side of the headrest and faces forward of the headrest, and the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone disposed on the front side of the headrest is greater than or equal to 3 cm.

[0137] For example, referring to Figure 9 or Figure 10, two speakers are symmetrically arranged on the front side of the headrest, both facing forward. For each speaker, the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone on the front side of the headrest is greater than or equal to 3 cm. As shown in Figure 9, the distance between the edge of the speaker 1 diaphragm and the pickup hole of error microphone 1 is r1, and the distance between the edge of the speaker 1 diaphragm and the pickup hole of error microphone 3 is r2, both r1 and r2 are greater than or equal to 3 cm. Similarly, as shown in Figure 10, the distance between the edge of the speaker 1 diaphragm and the pickup hole of error microphone 1 is r, where r is greater than or equal to 3 cm.

[0138] In some examples, a speaker (not shown) may also be placed at the center of the front side of the headrest, facing forward of the headrest, with the edge of the speaker's diaphragm at a distance greater than or equal to 3 cm from the pickup hole of each error microphone on the front side of the headrest.

[0139] Noise reduction can be achieved by placing speakers facing forward on the front of the headrest. These speakers output noise suppression signals to cancel out in-vehicle noise. The distance between the speakers and each error microphone located on the front of the headrest should be greater than a certain distance (e.g., 3cm) to avoid problems such as excessively large error microphone signal amplitude caused by speaker approach instability.

[0140] In some embodiments, the error microphone located on the front side of the headrest can be introduced into the aforementioned pickup hole or made of sound-permeable material to prevent the pickup channel of the error microphone from being blocked by the leather or sponge structure of the headrest, thus affecting the pickup of the error microphone. In some embodiments, the pickup hole can also be described as a sound guide hole.

[0141] (ii) The error microphone array is located on the front side of the seat.

[0142] This application provides a seat, including a headrest as described in any of the foregoing embodiments. The implementation principle and technical effects can be referred to the foregoing headrest embodiments, and will not be repeated here.

[0143] This application provides a seat comprising M+N error microphones, wherein M error microphones are located on the upper part of the left front region of the seat, and N error microphones are located on the upper part of the right front region of the seat, and the pickup holes of the M+N error microphones face the front of the seat. Wherein, N and M are both positive integers greater than 2.

[0144] It should be noted that the upper right front area of ​​the seat can refer to the front side of the seat near the head (or neck, or shoulder). N error microphones positioned on the front side of the seat remain stationary relative to the seat but can move with it.

[0145] In some embodiments, the seat includes a fifth microphone and a sixth microphone. The fifth microphone is located in the upper part of the left front area of ​​the seat, and the sixth microphone is located in the upper part of the right front area of ​​the seat. The horizontal distance between the pickup holes of the fifth microphone and the sixth microphone is set between 15cm and 30cm. The implementation principle and technical effects can be referred to the aforementioned headrest embodiment, and will not be repeated here.

[0146] In some embodiments, the seat includes a seventh microphone and an eighth microphone, both located in the upper part of the right front region or left front region of the seat, and the seventh and eighth microphones are two adjacent microphones. The distance between the pickup holes of the seventh and eighth microphones is set between 5cm and 15cm. The implementation principle and technical effects can be referred to the aforementioned headrest embodiment, and will not be repeated here.

[0147] In some embodiments, the seat further includes a speaker located on the upper front side of the seat and facing forward, wherein the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone disposed on the front side of the seat is greater than or equal to 3 cm. The implementation principle and technical effects can be referred to the aforementioned headrest embodiment, and will not be repeated here.

[0148] Based on the aforementioned seat or headrest, this application provides a noise reduction processing method. By sensing changes in the sound field and the desired control sound field, the active noise reduction strategy is dynamically adjusted to achieve better active noise reduction effects. The main idea is as follows: By analyzing signals collected by multiple error microphones, the weight values ​​corresponding to the signals of each error microphone in the noise reduction processing algorithm are dynamically adjusted. For example, the weight values ​​corresponding to the signals of each error microphone are adjusted according to the distance between the error microphone and the human ear; or the weight value corresponding to the signal of the interference error microphone is set to 0 (i.e., this signal does not participate in the noise reduction processing algorithm). A preset active noise reduction algorithm processes the signals participating in the calculation, generating control signals for noise cancellation in real time, thereby improving the noise reduction effect.

[0149] Interference with the error microphone can be caused by various factors, including wind noise, obstruction by heads, speech interference, friction noise, and slapping noise. For example, if the distance between the back of a person's head and the microphone is less than a certain threshold, the microphone is considered obstructed. Wind noise interference occurs when the microphone is affected by an open window or air conditioning vents. Speech interference occurs when the microphone receives a signal containing speech. Friction noise interference occurs when clothing rubs against the microphone. Slapping or knocking interference occurs when the microphone is struck or tapped.

[0150] The noise reduction method will be described in detail below with reference to several specific examples.

[0151] Figure 11 shows a flowchart of a noise reduction processing method. This noise reduction processing method can be applied to any noise reduction processing device, and the noise reduction processing device can be integrated into the vehicle controller; however, this embodiment does not limit this. For ease of understanding, the following description uses the noise reduction processing device as the execution subject. As shown in Figure 11, the noise reduction processing method includes:

[0152] S101. Receive at least one reference signal and signals acquired from N+M error microphones.

[0153] Where N and M are both positive integers greater than 2. The N+M error microphones can be error microphones installed in front of a headrest or a seat in the vehicle. The setup method for the N+M error microphones can be referred to the previous text.

[0154] After the vehicle starts or while it is in motion, the noise source acquisition device on the vehicle acquires a reference signal, and N+M error microphones acquire signals near the human ear. At the same time, the noise reduction processing device receives at least one reference signal from the noise source acquisition device, as well as signals acquired from the N+M error microphones.

[0155] The number of noise source acquisition devices is at least one, and correspondingly, the number of reference signals is at least one. This embodiment does not limit the number of noise source acquisition devices. For example, a noise source acquisition device can be placed near the engine to acquire signals from the vicinity of the engine. A noise source acquisition device can also be placed near the air conditioning vent to acquire signals from the vicinity of the air conditioning vent.

[0156] S102. Based on the signals acquired by at least P error microphones out of N+M error microphones, and at least one reference signal, generate a first noise suppression signal.

[0157] S103. Output the first noise suppression signal.

[0158] Among them, the signals from at least P error microphones are used for noise reduction processing, where P is a positive integer greater than 1 and less than or equal to N+M.

[0159] In some embodiments, P equals M+N, and the noise reduction processing device generates a first noise suppression signal based on the signals collected by N+M error microphones and at least one reference signal. That is, the signals collected by N+M error microphones all participate in the operation and processing of the noise reduction processing algorithm.

[0160] In some embodiments, P is less than M+N and greater than 1. The noise reduction processing device generates a first noise suppression signal based on the signals collected by P error microphones and at least one reference signal. That is, signals collected by some of the N+M error microphones participate in the operation and processing of the noise reduction processing algorithm.

[0161] The principle of the noise reduction algorithm in this embodiment will be explained in detail below with reference to Figure 12.

[0162] Referring to Figure 12, at a certain time n, the noise reduction processing device receives the original reference signal acquired by the noise source acquisition device. Original reference signal After processing by the filtering module, the filtered reference signal is obtained. Simultaneously, the noise reduction processing device receives signals from N+M error microphones. The noise reduction processing device updates the filter parameters using Formula 1.

[0163] In the formula, This represents the filter parameters at time n+1. Let Ω represent the filter parameters at time n, Ω represent the set of uninterrupted error microphones, and α represent the filter parameters at time n. m e represents the weight value corresponding to the signal (error signal) collected by the m-th error microphone in the set. m (n) represents the signal acquired by the m-th error microphone at time n, where m ranges from 1 to P. μ is the update factor, which takes values ​​from (0,1]. The larger the value, the larger the filter parameter update.

[0164] The following examples illustrate α m The settings will be explained.

[0165] In some examples, the signals collected by N+M error microphones all participate in the noise reduction algorithm, i.e., P equals N+M, where m can take values ​​from 1 to N+M, and α can be... m Set to 1.

[0166] In some examples, α m The value of α can be determined based on the distance between the m-th error microphone and the human ear. The smaller the distance value, the closer the m-th error microphone is to the human ear. m The larger the value of α, the further away the m-th error microphone is from the ear, and vice versa. m The smaller the value, the more reliable the signal collected by the error microphone is as it is closer to the ear. By increasing its weight, the accuracy of the algorithm can be improved, thereby enhancing the noise reduction effect.

[0167] In some examples, if it is determined that the m-th error microphone is interfered with (how to determine whether an error microphone is interfered with will be discussed later), α can be... m Setting it to 0 means that the m-th error microphone does not participate in the algorithm's calculation. By excluding error microphones that are not highly relevant to the algorithm, the accuracy of the algorithm's calculation is improved, thereby enhancing the noise reduction effect.

[0168] Referring again to Figure 12, the noise reduction processing device determines the latest parameters of the filter based on the aforementioned Formula 1 (e.g., ...). After that, the filter parameters are updated by control, and then the control signal y(n) at time n is obtained from the output of the filter. After the control signal y(n) is amplified by the power amplifier, the first noise suppression signal is obtained. Then, the first noise suppression signal is output through the speaker to achieve noise reduction.

[0169] The following describes how to determine P error microphones from N+M error microphones through several specific implementation methods.

[0170] In one possible implementation, the noise reduction processing device determines P interference-free error microphones from the N+M error microphones based on the signals collected by the N+M error microphones. These P interference-free error microphones include those not affected by wind noise, those not obstructed by heads, those not affected by voice noise, those not affected by friction noise, and those not affected by slapping noise, etc.

[0171] The following example illustrates how interference with error microphones can be determined: The noise reduction processing device removes target microphones that meet a first condition from N+M error microphones to identify P error microphones. The first condition includes at least one of the following:

[0172] 1) The decibel value corresponding to the attenuation of the signal collected by the target microphone in the first frequency band is greater than the first threshold.

[0173] For example, the first frequency band can be a frequency band from 500 to 20000 Hz. If the decibel value corresponding to the attenuation of the signal collected by the target microphone in the first frequency band is greater than the first threshold, it indicates that there is attenuation in the mid-to-high frequencies of the signal collected by the target microphone, and the target microphone can be considered to be blocked. For example, if it is blocked by a person's head, the target microphone can be eliminated.

[0174] 2) The energy value of the signal collected by the target microphone in the full frequency band or the second frequency band is greater than the second threshold.

[0175] For example, the second frequency band can be a frequency band from 1000 to 8000 Hz. If the energy value of the signal collected by the target microphone is greater than the second threshold in the full frequency band or the second frequency band, it indicates that there is a friction sound signal in the signal collected by the target microphone. It can be considered that the target microphone is interfered with by friction sound, such as the friction of clothing, and the target microphone can be eliminated.

[0176] 3) The correlation between the signal acquired by the target microphone and the signals acquired by other error microphones is less than the third threshold.

[0177] Signal correlation refers to the degree of similarity between two signals. If the correlation between the signal collected by the target microphone and the signals collected by other error microphones on the headrest or seat is less than the third threshold, it indicates that the target microphone is being interfered with and can be eliminated.

[0178] For example, referring to Figure 7a or Figure 7b, if the correlation between the signal collected by error microphone 1 and the signal collected by error microphone 2 is less than the third threshold, the correlation between the signal collected by error microphone 1 and the signal collected by error microphone 3 is less than the third threshold, and the correlation between the signal collected by error microphone 1 and the signal collected by error microphone 4 is less than the third threshold, it indicates that error microphone 1 is being interfered with, and error microphone 1 can be removed. The signal collected by error microphone 1 will not participate in the noise reduction processing algorithm.

[0179] The above example performs frequency domain analysis on the signals collected by N+M error microphones, and / or calculates the correlation between the signals collected by each pair of error microphones. Combined with the aforementioned first condition, this allows for the elimination of interfered error microphones, thus identifying the undisturbed error microphones among the N+M microphones. This prevents signals from interfered error microphones from participating in subsequent noise reduction processing, improving the accuracy of the noise reduction operation and ultimately enhancing the noise reduction effect.

[0180] In one possible implementation, the noise reduction processing device determines P undisturbed error microphones from N+M error microphones based on signals collected by sensors.

[0181] Optionally, the sensor includes at least one of an image sensor, an infrared sensor, and an ultrasonic sensor. The image sensor is the core component of the camera, which is typically positioned in front of the headrest / seat to capture images of the person's head, the headrest, and the seat. The infrared sensor is typically positioned in front of the headrest / seat to capture the relative position of the back of the head to the front of the headrest / seat. Both the infrared sensor and the ultrasonic sensor can also be positioned in front of the headrest / seat to capture the relative position of the person's head to the headrest / seat.

[0182] In one example of this embodiment, the noise reduction processing device determines the first relative positional relationship between the human head and the headrest or seat based on the signals collected by the sensor; based on the first relative positional relationship and the positional information of N+M error microphones on the headrest or seat, the target microphones that are blocked by the human head are removed from the N+M error microphones to determine P error microphones.

[0183] For example, taking an image sensor as the sensor and an error microphone positioned on the front of a headrest, the first relative positional relationship includes the positional information of the headrest area obscured by a person's head and the positional information of the headrest area not obscured by a person's head. If the positional information of the target microphone on the headrest overlaps with the positional information of the headrest area obscured by a person's head, it indicates that the target microphone is obscured by a person's head, and the target microphone can be eliminated.

[0184] For example, taking an infrared sensor as the primary sensor, with both the infrared sensor and the error microphone positioned on the front of the headrest, the first relative positional relationship between the back of the head and the front of the headrest can be obtained based on the signal collected by the infrared sensor. This first relative positional relationship includes the distance between the back of the head and the front of the headrest in a first direction, and the position information of the back of the head within the headrest area. The first direction is perpendicular to the plane of the front of the headrest. If the position information of the target microphone on the headrest overlaps with the position information of the back of the head within the headrest area (the back of the head obstructs the target microphone on the headrest), and the distance between the back of the head and the target microphone in the first direction is less than a certain distance threshold, it indicates that the target microphone is obstructed by the head, and the target microphone can be discarded.

[0185] In addition, based on the signals collected by the ultrasonic sensor, the first relative positional relationship between the back of the head and the front of the headrest can be obtained. The principle of the solution can be referred to the previous example, and will not be elaborated here.

[0186] The example above analyzes the signals collected by the sensors to determine the relative position of the person's head to N+M error microphones located on the front of the headrest or seat. Based on this relative position, error microphones obscured by the person's head are eliminated. This prevents signals from interfering error microphones from participating in subsequent noise reduction processing, improving the accuracy of the noise reduction process and thus enhancing the noise reduction effect.

[0187] In another example of this embodiment, the noise reduction processing device determines a second relative positional relationship between the human ear and the headrest or seat based on signals collected by sensors; based on the second relative positional relationship and the positional information of N+M error microphones on the headrest or seat, it removes target microphones from the N+M error microphones whose distance value from the human ear is greater than or equal to a preset distance value, thereby determining P error microphones. For example, the preset distance value can be set to 20cm.

[0188] In the example above, target microphones that are far from the ear are removed, even if those microphones are not interfered with. This reduces the number of signals involved in subsequent noise reduction processing, improving computational efficiency.

[0189] For example, taking an image sensor as the sensor and an error microphone positioned on the front of the headrest, the second relative positional relationship includes the positional information of the ear projected onto the headrest area. Based on the positional information of the ear projected onto the headrest area and the positional information of the target microphone on the headrest, the distance between the ear and each error microphone can be determined. If the distance between the ear and the target microphone is greater than or equal to a preset distance value, it indicates that the target microphone is too far from the ear, and the target microphone can be discarded. The signal collected by the target microphone will not participate in the subsequent noise reduction algorithm. Accordingly, the weight value corresponding to the signal collected by the target microphone can be set to 0. In addition, the positional information of the ear projected onto the headrest area can be obtained based on the signals collected by an infrared sensor or an ultrasonic sensor. The principle of this scheme can be referred to in this example and will not be elaborated here.

[0190] For example, taking N=4 as an example, referring to Figure 13a, the error microphones set on the front side of the headrest or seat include: a first microphone, a second microphone, a third microphone, and a fourth microphone, corresponding to error microphones 1, 2, 3, and 4 in Figure 13a, respectively. If the first distance value between the first microphone and the left ear is less than a preset distance value, the second distance value between the second microphone and the right ear is less than a preset distance value, the third distance value between the third microphone and the left ear is greater than or equal to a preset distance value, and the fourth distance value between the fourth microphone and the right ear is greater than or equal to a preset distance value, then the third microphone and the fourth microphone are discarded. As shown in Figure 13a, since the distance value x3 between the error microphone 3 and the left ear is greater than a preset distance value, and the distance value x4 between the error microphone 4 and the right ear is greater than a preset distance value, error microphones 3 and 4 can be discarded. That is, the signals collected by error microphones 3 and 4 do not participate in the subsequent noise reduction processing algorithm. Accordingly, referring to Formula 1, the weight values ​​corresponding to the signals collected by error microphones 1 and 2 can be set to 1, and the weight values ​​corresponding to the signals collected by error microphones 3 and 4 can be set to 0.

[0191] The above example analyzes the signals collected by the sensors to determine the relative positional relationship between the human ear and N+M error microphones placed on the front of the headrest or seat. Based on this relative positional relationship, error microphones that are far away from the human ear can be eliminated, and the signals collected by error microphones that are closer to the human ear can be used for noise reduction calculation to achieve adaptive single-area noise reduction, which can improve the noise reduction effect.

[0192] In some embodiments, the noise reduction processing device determines a second relative positional relationship between the human ear and the headrest or seat based on signals collected by sensors; determines weight values ​​corresponding to the signals collected by at least P error microphones based on the second relative positional relationship and the positional information of at least P error microphones on the headrest or seat; and generates a first noise suppression signal based on the weight values ​​corresponding to the signals collected by at least P error microphones, the signals collected by at least P error microphones, and at least one reference signal.

[0193] In this embodiment, after determining the distance between the human ear and each error microphone, the noise reduction processing device can set the weight value corresponding to the signal collected by each error microphone based on the distance between the human ear and each error microphone. The weight value can be set based on the principle that the larger the distance, the smaller the weight value. For example, referring to Figure 13b, since the left ear is closer to error microphone 1 and the right ear is closer to error microphone 2, the weight values ​​corresponding to the signals collected by error microphones 1 and 2 can be increased, and correspondingly, the weight values ​​corresponding to the signals collected by error microphones 3 and 4 can be decreased. For example, the weight value corresponding to the signals collected by error microphones 1 and 2 can be set to 2, and the weight value corresponding to the signals collected by error microphones 3 and 4 can be set to 0.5.

[0194] Compared to a noise reduction scheme where the weight of each error microphone's signal is set to 1, although both are full-area noise reduction schemes (i.e., all signals from error microphones participate in the calculation), the noise reduction effect can be further improved by fine-tuning the weight of each error microphone's signal.

[0195] In one possible implementation, the noise reduction processing device determines P undisturbed error microphones from the N+M error microphones based on the signals collected by the N+M error microphones and the sensor parameter signals. The implementation principle and effect are the same as those described in the two aforementioned implementations, and will not be elaborated upon here.

[0196] The noise reduction processing algorithms shown in the foregoing embodiments include the following data participating in the noise reduction processing algorithm operation: at least one filtered reference signal, and all or part of the signals collected by the error microphone, as shown in Figure 12.

[0197] Considering that the error microphone involved in noise reduction is a certain distance from the human ear, the signal collected by the error microphone is not the signal collected at the human ear. If virtual microphone technology is used, that is, the signal at the human ear (also known as the virtual microphone signal) is estimated by using the signal collected by the error microphone (also known as the real microphone), and the noise reduction operation is performed using the estimated signal at the human ear, the noise reduction effect can be further improved.

[0198] Figure 14 shows a flowchart of another noise reduction processing method, which can be applied to any noise reduction processing device. As shown in Figure 14, the noise reduction processing method includes:

[0199] S201. Receive at least one reference signal and signals acquired from N+M error microphones.

[0200] S202. Based on the signals collected by at least P error microphones out of N+M error microphones, generate the signal of a virtual microphone near the human ear.

[0201] The noise reduction processing device first identifies at least P uninterrupted error microphones from N+M error microphones; then, based on the signals collected by the at least P error microphones, it generates signals from virtual microphones near the human ear. These virtual microphone signals near the human ear include signals from virtual microphones near the left ear and virtual microphones near the right ear.

[0202] S203. Generate a first noise suppression signal based on the signal from a virtual microphone near the ear and at least one reference signal.

[0203] S204. Output the first noise suppression signal.

[0204] The principle of the noise reduction algorithm in this embodiment will be explained in detail below with reference to Figure 15.

[0205] Referring to Figure 15, at a certain time n, the noise reduction processing device receives the original reference signal acquired by the noise source acquisition device. Original reference signal After processing by the filtering module, the filtered reference signal is obtained. Simultaneously, the noise reduction processing device receives signals from N+M error microphones and, based on signals from at least P of these error microphones, estimates the signals from the virtual microphones near the left and right ears. The noise reduction processing device then updates the filter parameters using Equation 2.

[0206] In the formula, This represents the filter parameters at time n+1. Let e1 represent the filter parameters at time n, where μ is the update factor. ′ (n) represents the signal from the virtual microphone near the left ear, α1 represents the weight value of the signal from the virtual microphone near the left ear, and e2 represents the weight value of the signal from the virtual microphone near the left ear. ′ (n) represents the signal from the virtual microphone near the right ear, and α2 represents the weight value of the signal from the virtual microphone near the right ear.

[0207] In some embodiments, both α1 and α2 can be set to 1.

[0208] In some embodiments, the noise reduction processing device includes preset signal estimation-related models, and there may be multiple such models. Based on the estimated location of the virtual microphone, the models can be divided into two main categories: a first model for estimating the virtual microphone signal near the left ear, and a second model for estimating the virtual microphone signal near the right ear.

[0209] For example, Figures 16a and 16b illustrate the principle for estimating the signal from a virtual microphone near the left ear. Signals collected by four real microphones (i.e., error microphones) on the headrest are input into the first model. After processing by the first model, the signal from the virtual microphone 5 near the left ear is obtained. The signal e1 of the virtual microphone near the left ear can be determined using Formula 3. ′ (n):

[0210] e1 ′ (n)=∑ m∈Ω e m (n)h 1m (n) Formula 3

[0211] In the formula, Ω represents the set of error microphones that have not been interfered with, and e m (n) represents the signal acquired by the m-th error microphone at time n, h 1m (n) represents the parameters of the transmission path filter from the m-th error microphone to the virtual microphone near the left ear in the first model at time n. In Figure 16a or Figure 16b, four error microphones participate in signal estimation, where m ranges from 1 to 4, and h... 11 (n) represents the parameters of the transmission path filter from error microphone 1 to virtual microphone 5 in the first model at time n, h 12 (n) represents the parameters of the transmission path filter from error microphone 2 to virtual microphone 5 in the first model at time n, h 13 (n) represents the parameters of the transmission path filter from error microphone 3 to virtual microphone 5 in the first model at time n, h 14 (n) represents the parameters of the transmission path filter from error microphone 4 to virtual microphone 5 in the first model at time n.

[0212] Similarly, the signals collected by the four real microphones on the headrest are input into the second model. After processing by the second model, the signal from the virtual microphone near the right ear is obtained. The signal e2 of the virtual microphone near the right ear can be determined using Formula 4. ′ (n):

[0213] e2 ′ (n)=∑ m∈Ω e m (n)h2m (n) Formula 4

[0214] In the formula, Ω represents the set of error microphones that have not been interfered with, and e m (n) represents the signal acquired by the m-th error microphone at time n, h 2m (n) represents the parameters of the transmission path filter from the m-th error microphone to the right ear virtual microphone in the second model at time n.

[0215] Referring again to Figure 15, the noise reduction processing device determines the latest parameters of the filter (e.g., based on the aforementioned Formulas 2, 3, and 4) using formulas 2, 3, and 4. After that, the filter parameters are updated by control, and then the control signal y(n) at time n is obtained from the output of the filter. After the control signal y(n) is amplified by the power amplifier, the first noise suppression signal is obtained. Then, the first noise suppression signal is output through the speaker to achieve noise reduction.

[0216] Furthermore, depending on the number and location of the error microphones involved in signal estimation, the first model may include multiple models, such as Model 1, Model 2, etc. For example, the input to Model 1 includes signals collected by real microphones 1 to 3 in Figure 16a or Figure 16b, and Model 1 estimates the signal of the virtual microphone 5 near the left ear based on the signals collected by real microphones 1 to 3. The input to Model 2 includes signals collected by real microphones 1, 2, and 4 in Figure 16a or Figure 16b, and Model 2 estimates the signal of the virtual microphone 5 near the left ear based on the signals collected by real microphones 1, 2, and 4.

[0217] Similarly, depending on the number and location of the error microphones involved in signal estimation, the second model can include multiple models, such as Model 3, Model 4, etc. For example, the input to Model 3 includes signals collected by real microphones 1 to 3 in Figure 16a or Figure 16b, and Model 3 estimates the signal of a virtual microphone (not shown) near the right ear based on the signals collected by real microphones 1 to 3. The input to Model 4 includes signals collected by real microphones 2, 3, and 4 in Figure 16a or Figure 16b, and Model 4 estimates the signal of a virtual microphone near the right ear based on the signals collected by real microphones 2, 3, and 4.

[0218] It should be noted that all the models pre-installed in the noise reduction processing device are trained based on a large amount of training data, and the data used by the models in practical applications can also be used to update the models. In addition, this embodiment does not limit the structure of the models.

[0219] Based on the foregoing description, in some embodiments, after the noise reduction processing device determines at least P uninterrupted error microphones from N+M error microphones, it needs to select two models corresponding to the at least P error microphones from a set of preset models, denoted as the first model and the second model, based on the position of the at least P error microphones on the headrest or seat. The first model is used to estimate the signal of the virtual microphone near the left ear, and the second model is used to estimate the signal of the virtual microphone near the right ear. Based on this, the noise reduction processing device can perform the following steps:

[0220] Signals collected by at least P error microphones are input into a pre-trained first model corresponding to at least P error microphones. After processing by the first model, the signal from the virtual microphone near the left ear is obtained; and...

[0221] The signals collected by at least P error microphones are input into the pre-trained second model corresponding to at least P error microphones. After processing by the second model, the signal from the virtual microphone near the right ear is obtained.

[0222] The noise reduction processing scheme shown in the above embodiments uses the signals collected by all or part of the error microphones on the seat or headrest to estimate the signal of the virtual microphone near the ear, and updates the filter parameters based on the signal of the virtual microphone near the ear and at least one reference signal, and then generates a first noise suppression signal based on the updated filter parameters.

[0223] Compared to Scheme 1, where the left error microphone on the headrest or seat only estimates the left virtual microphone signal and the right error microphone only estimates the right virtual microphone signal, and Scheme 2, where one error microphone on each side of the headrest or seat estimates the left / right virtual microphone signal, this scheme has higher signal estimation accuracy and better robustness because more error microphones participate in signal estimation simultaneously (e.g., two error microphones on the left and two on the right).

[0224] Table 1 shows the noise reduction levels for the left and right ears for different schemes.

[0225] This application also proposes a noise reduction processing method. When wind noise interference or voice interference is detected, the filter parameters can be left unchanged, and the noise suppression signal of the previous moment can be maintained to avoid the noise reduction effect caused by wind noise or voice interference.

[0226] Figure 17 shows a flowchart of another noise reduction processing method provided in an embodiment of this application. As shown in Figure 17, this noise reduction processing method can be applied to any noise reduction processing device, and the method includes:

[0227] S301. At the first moment, receive at least one reference signal and signals collected from N+M error microphones.

[0228] S302. Generate a first noise suppression signal based on signals acquired by at least P error microphones out of N+M error microphones, and at least one reference signal.

[0229] S303. Output the first noise suppression signal.

[0230] S304. At the second moment, if the second condition is met, maintain the output of the first noise suppression signal.

[0231] In this case, the second moment is later than the first moment, such as the second moment being the moment following the first moment.

[0232] The second condition includes at least one of the following: detecting that at least one error microphone is being blown by air, or detecting that the signal from at least one error microphone contains a human voice signal.

[0233] In some embodiments, the noise reduction processing device, based on signals collected by N+M error microphones, detects that at least one error microphone is being blown by wind. It can then maintain the previous noise suppression signal output through the speaker without updating the current filter parameters. This avoids a decrease in noise reduction effect due to open windows or air conditioning drafts.

[0234] For example, if the energy value of the signal collected by the target error microphone in the third frequency band is greater than the fourth threshold, it indicates that the target microphone is being blown by wind, and it can be considered that the target microphone is subject to wind noise interference. The third frequency band can be a frequency band of 50 to 500 Hz.

[0235] Typically, the correlation between the signals collected by the two error microphones that are being blown by air is less than threshold 1 (poor correlation), the correlation between the signals collected by the error microphone that is being blown by air and the error microphone that is not being blown by air is less than threshold 2, and the correlation between the signals collected by the two error microphones that are not being blown by air is greater than threshold 3. Therefore, in some embodiments, it is also possible to determine whether at least one error microphone has been blown by air by analyzing the correlation between the signals collected by N+M error microphones.

[0236] In some embodiments, the noise reduction processing device, based on signals collected by N+M error microphones, detects that at least one error microphone contains a voice signal. It can then maintain the output of the noise suppression signal from the previous moment through the speaker without updating the current filter parameters. This avoids a decrease in noise reduction effectiveness due to voice interference.

[0237] For example, by extracting the speech signal and noise signal from the signal collected by the target microphone, if the energy value of the speech signal is greater than a certain threshold, or the energy value of the speech signal is greater than a certain threshold of the energy value of the noise signal, or the ratio of the energy value of the speech signal to the energy value of the noise signal is greater than a certain threshold (exceeding this threshold will affect the noise reduction performance), it indicates that the signal of the target microphone contains a voice signal.

[0238] This application provides a noise reduction processing device, including: a processor and a memory; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory, causing the noise reduction processing device to perform the steps of any of the foregoing method embodiments. The implementation principle and technical effects can be referred to the relevant embodiments, which will not be repeated here.

[0239] Figure 18 shows a schematic diagram of a noise reduction processing device. As shown in Figure 18, the noise reduction processing device includes a processor 1801, a communication line 1804, and at least one communication interface (in Figure 18, communication interface 1803 is used as an example for illustration).

[0240] The processor 1801 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application.

[0241] Communication line 1804 may include circuitry for transmitting information between the aforementioned components.

[0242] The communication interface 1803 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, wireless local area networks (WLAN), etc.

[0243] In some embodiments, the noise reduction processing apparatus may further include a memory 1802.

[0244] The memory 1802 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory may exist independently and be connected to the processor via communication line 1804. The memory may also be integrated with the processor.

[0245] The memory 1802 stores computer execution instructions for implementing the scheme of this application, and the processor 1801 controls the execution. The processor 1801 executes the computer execution instructions stored in the memory 1802, thereby implementing the noise reduction processing method provided in the embodiments of this application.

[0246] The computer execution instructions in the embodiments of this application may also be referred to as application code, and the embodiments of this application do not specifically limit this.

[0247] As an example, processor 1801 may include one or more CPUs.

[0248] As an example, a noise reduction processing device may include multiple processors. Each processor may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (such as computer program instructions).

[0249] Based on the foregoing embodiments, this application also provides a vehicle, including a seat of any of the foregoing embodiments and a noise reduction processing device of any of the foregoing embodiments. The noise reduction processing device is used to perform the steps of any of the foregoing method embodiments, and its implementation principle and technical effects can be referred to the relevant embodiments, which will not be repeated here.

[0250] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of any of the foregoing method embodiments. The implementation principle and technical effects can be referred to the relevant embodiments, which will not be repeated here.

[0251] The methods described in the above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. If implemented in software, the functionality can be stored as one or more instructions or code on or transmitted on a computer-readable medium. A computer-readable medium can include computer storage media and communication media, and can also include any medium that can transfer a computer program from one place to another. A storage medium can be any target medium accessible by a computer.

[0252] In one possible implementation, a computer-readable medium may include RAM, ROM, compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage or other magnetic storage devices, or any other medium targeted to carry or to store the required program code in the form of instructions or data structures, and accessible by a computer. Furthermore, any connection is appropriately referred to as a computer-readable medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. As used herein, disks and optical discs include optical discs, laser discs, optical discs, Digital Versatile Discs (DVDs), floppy disks, and Blu-ray discs, where disks typically reproduce data magnetically, while optical discs optically reproduce data using lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0253] This application provides a chip system or chip, including at least one processor and a communication interface. The communication interface and at least one processor are interconnected via a line. The at least one processor is used to run a computer program or instructions to perform the steps of any of the foregoing method embodiments. The implementation principle and technical effects can be referred to the relevant embodiments, which will not be repeated here.

[0254] Figure 19 shows a schematic diagram of a chip structure. As shown in Figure 19, the chip 1900 includes one or more (including two) processors 1920 and a communication interface 1930.

[0255] In some embodiments, memory 1940 stores the following elements: executable modules or data structures, or a subset of executable modules or data structures, or an extended set of executable modules or data structures.

[0256] In this embodiment, memory 1940 may include read-only memory and random access memory, and provides instructions and data to processor 1920. A portion of memory 1940 may also include non-volatile random access memory (NVRAM).

[0257] In this embodiment, the memory 1940, the communication interface 1930, and the memory 1940 are coupled together via a bus system 1910. The bus system 1910 may include a data bus, as well as a power bus, a control bus, and a status signal bus. For ease of description, all buses are labeled as bus system 1910 in Figure 19.

[0258] The methods described in the embodiments of this application can be applied to, or implemented by, processor 1920. Processor 1920 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above methods can be completed by integrated logic circuits in the hardware of processor 1920 or by instructions in software form. Processor 1920 may be a general-purpose processor (e.g., a microprocessor or conventional processor), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates, transistor logic devices, or discrete hardware components. Processor 1920 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0259] This application provides a computer program product, including a computer program. When the computer program is run, it causes the computer to perform the steps of any of the foregoing method embodiments. The implementation principle and technical effects can be referred to the relevant embodiments, which will not be repeated here.

[0260] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A noise reduction processing method, characterized in that, include: Receive at least one reference signal and signals acquired from N+M error microphones, where N and M are both positive integers; A first noise suppression signal is generated based on the signals collected by at least P error microphones out of the N+M error microphones and the at least one reference signal; the signals from the at least P error microphones are used for noise reduction processing, where P is a positive integer greater than or equal to 1 and less than or equal to N+M. Output the first noise suppression signal.

2. The method according to claim 1, characterized in that, The method further includes: Based on the signals collected by the N+M error microphones, and at least one of the signals collected by the sensor, at least P error microphones that have not been interfered with are determined from the N+M error microphones.

3. The method according to claim 2, characterized in that, Based on the signals collected by the N+M error microphones, determining at least P error microphones that have not been interfered with from the N+M error microphones includes: From the N+M error microphones, target microphones that meet a first condition are removed to determine the at least P error microphones; the first condition includes at least one of the following: The signal acquired by the target microphone has a decibel value corresponding to attenuation in the first frequency band that is greater than a first threshold. The energy value of the signal collected by the target microphone in the full frequency band or the second frequency band is greater than the second threshold. The correlation between the signal acquired by the target microphone and the signals acquired by other error microphones is less than a third threshold.

4. The method according to claim 2, characterized in that, Based on the signals acquired by the sensor, determining at least P undisturbed error microphones from the N+M error microphones includes: Based on the signals collected by the sensor, the first relative positional relationship between the human head and the headrest or seat is determined; Based on the first relative positional relationship and the positional information of the N+M error microphones on the headrest or the seat, the target microphones that are obscured by the head are removed from the N+M error microphones to determine the at least P error microphones.

5. The method according to claim 2, characterized in that, Based on the signals acquired by the sensor, determining at least P undisturbed error microphones from the N+M error microphones includes: Based on the signals collected by the sensor, a second relative positional relationship between the human ear and the headrest or seat is determined; Based on the second relative position relationship and the position information of the N+M error microphones on the headrest or the seat, target microphones whose distance value from the human ear is greater than or equal to a preset distance value are removed from the N+M error microphones to determine the at least P error microphones.

6. The method according to claim 5, characterized in that, The N+M error microphones include: a first microphone, a second microphone, a third microphone, and a fourth microphone; The step of removing target microphones from the N+M error microphones that are at a distance greater than a preset distance from the human ear, based on the second relative positional relationship and the positional information of the N+M error microphones on the headrest or the seat, includes: If the first distance between the first microphone and the left ear is less than the preset distance value, the second distance between the second microphone and the right ear is less than the preset distance value, the third distance between the third microphone and the left ear is greater than or equal to the preset distance value, and the fourth distance between the fourth microphone and the right ear is greater than or equal to the preset distance value, then the third microphone and the fourth microphone are discarded.

7. The method according to any one of claims 1 to 6, characterized in that, The generation of the first noise suppression signal based on signals acquired by at least P error microphones out of the N+M error microphones, and the at least one reference signal, includes: Based on signals collected by sensors, the second relative positional relationship between the human ear and the headrest or seat is determined; Based on the second relative position relationship and the position information of the at least P error microphones on the headrest or the seat, the weight values ​​corresponding to the signals collected by the at least P error microphones are determined. The first noise suppression signal is generated based on the weight values ​​corresponding to the signals collected by the at least P error microphones, the signals collected by the at least P error microphones, and the at least one reference signal.

8. The method according to any one of claims 2 to 7, characterized in that, The sensor includes at least one of an image sensor, an infrared sensor, and an ultrasonic sensor.

9. The method according to any one of claims 1 to 8, characterized in that, The generation of the first noise suppression signal based on signals acquired by at least P error microphones out of the N+M error microphones, and the at least one reference signal, includes: Based on the signals collected by the at least P error microphones, a signal from a virtual microphone near the human ear is generated; The first noise suppression signal is generated based on the signal from the virtual microphone and the at least one reference signal.

10. The method according to claim 9, characterized in that, The step of generating a signal from a virtual microphone near the human ear based on signals collected by the at least P error microphones includes: The signals collected by the at least P error microphones are input into the pre-trained first model corresponding to the at least P error microphones. After processing by the first model, the signal from the virtual microphone near the left ear is obtained; and... The signals collected by the at least P error microphones are input into the pre-trained second model corresponding to the at least P error microphones. After processing by the second model, the signal from the virtual microphone near the right ear is obtained.

11. The method according to any one of claims 1 to 10, characterized in that, The receiving of at least one reference signal and signals acquired from N+M error microphones includes: At the first moment, receive the at least one reference signal and the signals collected from the N+M error microphones; The method further includes: At the second moment, if the second condition is met, continue to output the first noise suppression signal; The second time point is later than the first time point, and the second condition includes at least one of the following: detecting that at least one error microphone is being blown by air, or detecting that there is a human voice signal from at least one error microphone.

12. A headrest, characterized in that, include: M+N error microphones, with M error microphones located in the left front area of ​​the headrest and N error microphones located in the right front area of ​​the headrest, and the pickup holes of the M+N error microphones facing the front of the headrest; Both N and M are positive integers.

13. The headrest according to claim 12, characterized in that, The headrest includes a first microphone and a second microphone. The first microphone is located in the left front region of the headrest, and the second microphone is located in the right front region of the headrest. The horizontal distance between the pickup holes of the first microphone and the second microphone is set between 15cm and 30cm.

14. The headrest according to claim 12 or 13, characterized in that, The headrest includes a third microphone and a fourth microphone. The third microphone and the fourth microphone are both located in the left front area or the right front area of ​​the headrest, and the third microphone and the fourth microphone are two adjacent microphones. The distance between the pickup hole of the third microphone and the pickup hole of the fourth microphone is set between 5cm and 15cm.

15. The headrest according to any one of claims 12 to 14, characterized in that, The headrest also includes a speaker located on the front side of the headrest and facing forward, wherein the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone is greater than or equal to 3 cm.

16. A type of seat, characterized in that, Including the headrest as described in any one of claims 12 to 15; or, Includes: M+N error microphones, M error microphones are located in the upper part of the left front area of ​​the seat, N error microphones are located in the upper part of the right front area of ​​the seat, and the pickup holes of the M+N error microphones face the front of the seat; Both N and M are positive integers.

17. The seat according to claim 16, characterized in that, The seat also includes a speaker located on the upper front side of the seat and facing forward of the seat, wherein the distance between the edge of the speaker diaphragm and the pickup hole of each error microphone is greater than or equal to 3 cm.

18. A noise reduction processing device, characterized in that, include: Processor and memory; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the noise reduction processing device to perform the method as described in any one of claims 1 to 11.

19. A vehicle, characterized in that, include: The seat as claimed in claim 16 or 17, and the noise reduction processing device as claimed in claim 18.

20. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 11.

21. A chip system, characterized in that, It includes at least one processor and a communication interface, the communication interface and the at least one processor being interconnected via a line, the at least one processor being used to run a computer program or instructions to perform the method as described in any one of claims 1 to 11.

22. A computer program product, characterized in that, Includes a computer program that, when run, causes a computer to perform the method as described in any one of claims 1 to 11.

Images