earphone

Abstract

The present application provides an earphone, comprising: a microphone, for collecting a first sound signal, the first sound signal comprising a first type of signal component and a second type of signal component; a spatial domain filter, connected to the microphone, for filtering the first sound signal to obtain a second sound signal; a noise reduction filter, connected to the spatial domain filter, for receiving the second sound signal and determining a noise reduction signal corresponding to the second sound signal; a speaker, connected to the noise reduction filter, for playing the noise reduction signal; wherein, when the earphone uses a first noise reduction mode, the spatial domain filter performs a first filtering process on the first sound signal, so that the second type of signal component passes through and the first type of signal component is cut off, and the first noise reduction mode is an active noise reduction mode for resisting wind noise; when the earphone uses a second noise reduction mode, the spatial domain filter performs a second filtering process on the first sound signal, so that the first type of signal component passes through and the second type of signal component is cut off, and the second noise reduction mode is a selective transparency mode.

Description

Technical Field

[0001] The present application relates to the technical field of electronic equipment, and in particular to a headset. Background Art

[0002] In recent years, earphones are used more and more frequently in people's daily life. In order to meet the diverse needs of users, earphones can have different modes. For example, some earphones can include active noise reduction mode, transparent mode, normal mode, etc.

[0003] However, there are still some problems with the existing modes of headphones. On the one hand, the existing active noise reduction mode and transparency mode have the problem of poor noise reduction effect, which affects the user experience. For example, the active noise reduction mode is affected by wind noise, resulting in wind noise that would not be heard by the human ear after filtering by the headphones, generating inverted "wind noise", making the human ear hear more obvious wind noise. Furthermore, the transparency mode abandons the active noise reduction function and performs distortion compensation on all sounds collected by the microphone (including ambient noise that the user is not interested in), causing the human ear to hear more obvious sound signals of no interest, and so on.

[0004] On the other hand, the signal processing logic corresponding to the active noise reduction mode and the transparency mode is completely different. When the headphones use different modes, different filtering systems are required to process the sound signals collected by the microphone. As a result, when the headphones switch between modes, the filtering system used must also be switched accordingly. Therefore, the algorithm design cost of existing headphones is relatively high, and there is no plan to explore the possibility of organically combining the two "opposite functions" of noise reduction and transparency. Summary of the invention

[0005] In view of this, the present application is committed to providing a headset that can avoid the impact of poor noise reduction effect of existing modes of headsets, improve user experience, and use a set of filtering systems to achieve filtering processing in different modes, which can significantly reduce the design cost of acoustic algorithms.

[0006] The earphones provided in the present application include: a microphone for collecting a first sound signal, wherein the first sound signal includes a first type of signal component and a second type of signal component; a spatial domain filter connected to the microphone and used to filter the first sound signal to obtain a second sound signal; a noise reduction filter connected to the spatial domain filter and used to receive the second sound signal and determine a noise reduction signal corresponding to the second sound signal; and a speaker connected to the noise reduction filter and used to play the noise reduction signal; wherein, when the earphones use a first noise reduction mode, the spatial domain filter is used to perform a first filtering process on the first sound signal, so that the second type of signal component passes and the first type of signal component is cut off, and the first noise reduction mode is an active noise reduction mode for resisting wind noise; when the earphones use a second noise reduction mode, the spatial domain filter is used to perform a second filtering process on the first sound signal, so that the first type of signal component passes and the second type of signal component is cut off, and the second noise reduction mode is a selective transparency mode.

[0007] Optionally, when the earphone uses the first noise reduction mode, the first type of signal component includes a pneumatic noise signal.

[0008] Optionally, the spatial domain filter includes a determination unit, a classification unit and a filtering unit, the determination unit is connected to the microphone, the classification unit is connected to the determination unit, and the filtering unit is connected to the classification unit, and the spatial domain filter is used to perform a first filtering process on the first sound signal, including: the determination unit is used to determine the sound source direction of the second type of signal component; the classification unit is used to perform binary classification processing on the first sound signal according to the sound source direction of the second type of signal component to obtain two signal sets; the filtering unit is used to perform fidelity processing on the signal set to which the second type of signal component belongs, and perform distortion processing on the signal set to which the first type of signal component belongs to obtain the second sound signal.

[0009] Optionally, the filtering unit performs distortion processing on the signal set to which the first type of signal component belongs, including: the filtering unit performs amplitude zeroing processing on the signal set to which the first type of signal component belongs; or the filtering unit performs phase shift processing on the signal set to which the first type of signal component belongs, so that there is no intersection in the time domain between the signal set to which the first type of signal component belongs and the signal set to which the second type of signal component belongs.

[0010] Optionally, when the earphone uses the second noise reduction mode, the first type of signal component includes an unidirectional fluctuating noise signal.

[0011] Optionally, the spatial domain filter includes a determination unit, a classification unit and a filtering unit, the determination unit is connected to the microphone, the classification unit is connected to the determination unit, and the filtering unit is connected to the classification unit, and the spatial domain filter is used to perform a second filtering process on the first sound signal, including: the determination unit is used to determine the sound source direction of the second type of signal component; the classification unit is used to perform binary classification processing on the first sound signal according to the sound source direction of the second type of signal component to obtain two signal sets; the filtering unit is used to perform fidelity processing on the signal set to which the first type of signal component belongs, and perform distortion processing on the signal set to which the second type of signal component belongs to obtain the second sound signal.

[0012] Optionally, the filtering unit performs distortion processing on the signal set to which the second-category signal component belongs, including: the filtering unit performs amplitude zeroing processing on the signal set to which the second-category signal component belongs; or the filtering unit performs phase shift processing on the signal set to which the second-category signal component belongs, so that the signal set to which the second-category signal component belongs and the signal set to which the first-category signal component belongs have no intersection in the time domain.

[0013] Optionally, the classification unit includes: a construction module, used to construct multiple basis function signals pointing to the sound source direction of the second type of signal component; wherein the multiple basis function signals are linearly independent of each other; a convolution module, used to convolve the first sound signal with the multiple basis function signals respectively, and determine the probability of the existence of the second type of signal component in each signal point in the first sound signal based on the convolution result; an estimation module, used to estimate the first type of signal component in each signal point according to the probability; and a cropping module, used to crop each signal point into the corresponding first type of signal component and the second type of signal component to obtain the two signal sets.

[0014] Optionally, the estimation module is used to estimate the first type of signal components in each signal point according to the probability, including: the estimation module is used to identify the signal points whose probability is less than a threshold as purely random points, and count all the purely random points to estimate the first type of signal components in each signal point.

[0015] Optionally, the number of microphones is at least two, and the determination unit includes: a first calculation module, used to calculate the time difference between the second type of signal components appearing successively in the acquisition sequence of each microphone; and a second calculation module, used to calculate the sound source direction of the second type of signal components based on the time difference.

[0016] The earphones provided in the embodiment of the present application can use a set of filtering systems (including spatial filters and noise reduction filters) to achieve different noise reduction modes, without designing two sets of circuits and switching the circuits. The filtering system can perform different filtering processes on the sound signals collected by the microphone according to the different noise reduction modes used by the earphones, ensuring that different signal components in the sound signal can be selectively passed based on the current noise reduction mode. In the active noise reduction mode, the wind noise that the active noise reduction system is forced to introduce into the human ear is effectively blocked. In the transparent mode, the environmental noise that the transparent system is forced to compensate for is removed. Both modes can optimize the noise reduction effect of the earphones and enhance the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the structure of the earphones provided in an embodiment of the present application.

[0018] Figure 2 A schematic diagram of the structure of the spatial domain filter provided in an embodiment of the present application.

[0019] Figure 3 A flowchart of filtering processing provided in one embodiment of the present application.

[0020] Figure 4 A flowchart of filtering processing provided in another embodiment of the present application.

[0021] Figure 5 A schematic diagram of the structure of the classification unit provided in the embodiment of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0023] It should be understood that, in the present application, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying an order in terms of space, time, logic, etc.

[0024] It should be understood that in the description of this application, unless otherwise clearly specified and limited, terms such as "connection" should be understood in a broad sense. For example, it can be a fixed connection or a detachable connection; it can be a mechanical connection or an electrical connection. For those skilled in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0025] In recent years, the frequency of earphones in people's daily life has increased. With the diversification of user needs, the update rate of earphones has also become more frequent, for example, the emergence of Bluetooth earphones and active noise reduction earphones. In addition, in order to meet the diverse needs of users, earphones can also have different modes. For example, some earphones can include active noise reduction mode, transparent mode, normal mode, etc.

[0026] As the name suggests, the active noise reduction mode can be used to perform active noise reduction processing on the noise signal collected by the microphone. For example, a noise reduction signal with the same frequency, equal amplitude, and opposite phase as the noise signal collected by the microphone is generated through a noise reduction filter to offset the noise signal collected by the microphone and achieve active noise reduction.

[0027] Unlike the active noise reduction mode, the transparent mode is not used to reduce the noise of the surrounding environment. Specifically, in order to avoid the active noise reduction mode and the physical isolation of the headphones reducing the noise of the surrounding environment, resulting in the user not being able to hear the sounds of interest, which may contain useful voice information or environmental sound information (such as the voice of a companion talking, the honking of a car behind you on the road, etc.), the transparent mode can compensate for the distortion of the sound signal collected by the microphone and play the sound signal after distortion compensation to the user through the headphones.

[0028] The normal mode performs neither active noise reduction nor distortion compensation, that is, the normal mode does not process the sound signals collected by the microphone (except for the physical isolation produced by the earphones themselves). When the user uses the normal mode of the earphones, the earphones are no different from ordinary earphones used in people's daily lives.

[0029] Although currently earphones can have different modes to adapt to the diverse needs of users, there are still some problems with the existing modes of earphones.

[0030] On the one hand, the existing active noise reduction mode and transparent mode have the problem of poor noise reduction effect. For example, due to the influence of wind noise, the environmental noise that the user is not interested in is distorted and compensated, which affects the user experience. The following is a detailed description of this problem with two examples.

[0031] As an example, when a user uses the active noise reduction mode of headphones in some windy environments (for example, outdoors), the wind noise (a kind of aerodynamic noise) generated by the wind is everywhere. In addition to wind noise, there is also environmental noise (a kind of fluctuating noise). In this environment, the microphone (or feedforward microphone, reference microphone, etc.) used to collect noise signals in the headphones can collect wind noise and environmental noise at the same time, and then transmit the mixed sound signal of the collected wind noise and environmental noise to the noise reduction filter to generate a corresponding noise reduction signal. However, aerodynamic noise is different from fluctuating noise. Due to the particularity of the generation mechanism of aerodynamic noise based on forced convection motion of particles, aerodynamic noise itself does not propagate into the inside of the headphones, that is, the wind noise outside the headphones will not pass through the headphone structure to be heard by the human ear, but it is collected by the microphone and transmitted to the noise reduction filter to generate an inverted "wind noise", which makes the human ear hear more obvious wind noise, resulting in poor noise reduction effect and affecting the user experience.

[0032] As another example, when the user uses the transparent mode of the headset, the transparent mode abandons the active noise reduction function and adopts a method of compensating the distortion of all ambient sounds collected by the microphone (including sounds that the user is interested in and sounds that the user is not interested in), causing the human ear to hear more obvious sound signals that the user is not interested in, that is, the transparent mode cannot reduce the noise of sound signals that the user is not interested in, affecting the user's experience. In the embodiment of the present application, the sound that the user is interested in may refer to the sound signal that contains useful information for the user; the sound that the user is not interested in may refer to the sound signal that does not contain useful information for the user.

[0033] On the other hand, the signal processing logic corresponding to the active noise reduction mode and the transparency mode is completely different. When the headphones use different modes, different filtering systems are required to process the sound signals collected by the microphone. As a result, when the headphones switch between modes, the filtering system used must also be switched accordingly. Therefore, the algorithm design cost of existing headphones is relatively high. At present, there is no plan to explore the organic combination of the two "opposite functions" of noise reduction and transparency in the underlying filtering system architecture.

[0034] In view of this, an embodiment of the present application provides an earphone that can avoid the influence of poor noise reduction effect of existing modes of the earphone, improve user experience, and can use a set of filtering systems to achieve filtering processing in different modes, without the need to design two sets of circuits, and without the need to control circuit switching to achieve mode switching.

[0035] The embodiment of the present application does not specifically limit the type of earphones. For example, the earphones mentioned in the embodiment of the present application may be wired earphones or Bluetooth earphones; or the earphones mentioned in the embodiment of the present application may be in-ear earphones, semi-in-ear earphones, or head-mounted earphones, etc.

[0036] The earphones provided in the embodiment of the present application may have different modes, for example, they may include an active noise reduction mode for resisting wind noise, a selective transparency mode, etc. However, the embodiment of the present application is not limited thereto. For example, in addition to these two modes, the earphones provided in the embodiment of the present application may also include an existing active noise reduction mode, an existing transparency mode, a normal mode, etc. The earphones provided in the embodiment of the present application are described in detail below in conjunction with the accompanying drawings.

[0037] Figure 1 This is a schematic diagram of the structure of the earphone provided in the embodiment of the present application. Figure 1 As shown, the headset 100 may include a microphone 110 , a spatial domain filter 120 , a noise reduction filter 130 , and a speaker 140 .

[0038] The microphone 110 may also be referred to as a feedforward microphone or a reference microphone. The microphone 110 may be mounted on a housing of the earphone 100 to collect a first sound signal.

[0039] The first sound signal may refer to a mixed sound signal collected by the microphone 110, and the mixed sound signal may include multiple signal components. As an example, in an outdoor scene, the first sound signal collected by the microphone 110 may include multiple signal components such as the sound of a person speaking, the surrounding sound, and wind noise. As another example, in an indoor scene, the first sound signal collected by the microphone 110 may include multiple signal components such as the sound of a person speaking, the surrounding sound, and so on.

[0040] In the embodiment of the present application, the signal components contained in the first sound signal can be divided into a first type of signal components and a second type of signal components according to the signal components of the first sound signal collected by the microphone 110, that is, the first sound signal includes the first type of signal components and the second type of signal components. However, the embodiment of the present application is not limited to the first sound signal including only two types of signal components, and it can also include a third type of signal components, etc.

[0041] According to different classification standards, the first type of signal components in the first sound signal may be different in different classification standards, and the second type of signal components may also be different in different classification standards. For example, according to the noise generation mechanism, the first type of signal components and the second type of signal components contained in the first sound signal may be aerodynamic noise and wave noise respectively; according to the directivity of the sound source, the first type of signal components and the second type of signal components contained in the first sound signal may be divided into non-directional noise (abbreviated as non-directional noise) and directional noise (abbreviated as directional noise), etc. It can be seen that in the comprehensive classification standard, the first type of signal component may refer to aerodynamic noise or non-directional wave noise, among which aerodynamic noise is a type of non-directional noise due to its generation mechanism; the second type of signal component may refer to directional wave noise.

[0042] The embodiment of the present application does not specifically limit the type of microphone 110. For example, microphone 110 may be an omnidirectional microphone. If microphone 110 is an omnidirectional microphone, microphone 110 can collect sounds from any direction, that is, no matter at which angle the sound source is, the microphone can collect all sound signals from different directions with the same sensitivity. Alternatively, microphone 110 may be a unidirectional microphone, a directional microphone, etc.

[0043] The embodiment of the present application does not specifically limit the number of microphones 110. For example, in some embodiments, only one microphone 110 may be provided in the headset, such as an omnidirectional microphone. In some embodiments, at least two microphones 110 (or a microphone array) may be provided in the headset, such as multiple omnidirectional microphones or multiple unidirectional microphones.

[0044] The spatial filter 120 can be connected to the microphone 110 to filter the first sound signal collected by the microphone 110 to obtain a second sound signal. Depending on the mode used by the headset, the spatial filter 120 can perform different filtering processes on the first sound signal. For example, when the headset uses the first noise reduction mode (for example, the first noise reduction mode is an active noise reduction mode for wind noise), the spatial filter 120 can be used to perform a first filtering process on the first sound signal, so that the second type of signal component in the first sound signal passes and the first type of signal component, i.e., the aerodynamic noise signal, is cut off. When the headset uses the second noise reduction mode (for example, the second noise reduction mode is a selective transparent mode), the spatial filter 120 can be used to perform a second filtering process on the first sound signal, so that the first type of signal component in the first sound signal passes and the second type of signal component, i.e., the directional fluctuation noise signal (including conversations, bus voice announcements, whistles, and other sounds of interest to users) is cut off. The so-called "pass" and "cut off" here refer to passing or cutting off in the filter circuit, not passing or cutting off in the physical space. The different filtering processes performed by the spatial domain filter 120 for different modes of headphone use will be described in detail later, and will not be described in detail here.

[0045] The embodiment of the present application does not specifically limit the type of the spatial domain filter 120, as long as it can implement spatial domain filtering of the first sound signal. As an implementation method, the spatial domain filter 120 may refer to a beam former, which performs spatial domain filtering by beam forming and beam nulling. It should be understood that in some embodiments, spatial domain filtering can also be understood as spatial filtering or directional filtering.

[0046] The noise reduction filter 130 can be connected to the spatial domain filter 120 to receive the second sound signal sent by the spatial domain filter (i.e., the sound signal obtained after the spatial domain filter 120 filters the first sound signal) and determine the noise reduction signal corresponding to the second sound signal. The embodiment of the present application does not specifically limit the type and structure of the noise reduction filter 130. For example, the noise reduction filter 130 can be an all-pass filter, a band-pass filter, etc.

[0047] For the noise reduction filter 130, generally speaking, its noise reduction parameters are designed offline and fixed. In some cases, for example, the difference in the current usage mode may cause the optimal noise reduction parameters required by the noise reduction filter 130 to shift. Therefore, in some embodiments, the noise reduction filter 130 may adopt an adaptive filter. When the noise reduction filter 130 adopts an adaptive filter, the optimal noise reduction parameters of the noise reduction filter 130 can be corrected online by adaptive control during the user's use phase. Optionally, when the noise reduction filter 130 adopts an adaptive filter, an error microphone can also be set near the sound outlet of the earphone to collect the residual noise signal in the ear under noise reduction.

[0048] The speaker 140 may be connected to the noise reduction filter 130 to play the noise reduction signal sent by the noise reduction filter 130 .

[0049] The earphones provided in the embodiment of the present application can use a set of filtering systems (including spatial filters and noise reduction filters) to achieve different noise reduction modes, without designing two sets of circuits and switching the circuits. The filtering system can perform different filtering processes on the sound signals collected by the microphone according to the different noise reduction modes used by the earphones, ensuring that different signal components in the sound signal can be selectively passed based on the current noise reduction mode. In the active noise reduction mode, the wind noise that the active noise reduction system is forced to introduce into the human ear is effectively blocked. In the transparent mode, the environmental noise that the transparent system is forced to compensate for is removed. Both modes can optimize the noise reduction effect of the earphones and enhance the user experience.

[0050] Combine the following Figure 2 , the spatial filter 120 performs different filtering processes for different modes of earphone use in detail. Specifically, the first noise reduction mode is an active noise reduction mode for resisting wind noise, and the second noise reduction mode is a selective transparency mode as an example for introduction.

[0051] In the embodiment of the present application, when the first noise reduction mode is an active noise reduction mode for resisting wind noise, the first type of signal component may include aerodynamic noise signals, such as wind noise signals, and the second type of signal component may include directional fluctuating noise signals, such as the noisy sounds of passers-by, the clamor of broadcasting in the environment, etc. When the second noise reduction mode is a selective transparency mode, the first type of signal component may include non-directional fluctuating noise signals, such as the sound of swaying branches, the sound emitted by the central air-conditioning outlet on the ceiling, and other sounds that do not contain valuable information for the user, and the second type of signal component may include directional fluctuating sound signals, such as the sound of a companion talking, the sound of a whistle from a certain direction of the user, and other sounds that contain valuable information for the user. In real scenarios, non-directional sounds are usually worthless for human information interaction, and the sounds of interest in human information interaction are almost all directional.

[0052] See also Figure 2 , the spatial domain filter 120 may include a determination unit 122, a classification unit 124, and a filtering unit 126. The determination unit 122 may be connected to the microphone 110, the classification unit 124 may be connected to the determination unit 122, and the filtering unit 126 may be connected to the classification unit 124. In combination with the specific structure of the spatial domain filter 120, the first filtering process performed by the spatial domain filter 120 on the first sound signal in the first noise reduction mode and the second filtering process performed by the spatial domain filter 120 on the first sound signal in the second noise reduction mode are introduced below.

[0053] When the earphone uses the first noise reduction mode, the spatial domain filter 120 performs a first filtering process on the first sound signal, which may include: Figure 3 Steps shown.

[0054] Specifically, in step S310 , the determination unit 122 determines the sound source direction of the second type signal component.

[0055] As described above, when the first noise reduction mode is an active noise reduction mode for resisting wind noise, the second type of signal component may include, for example, directional environmental noise such as the noisy sounds of passers-by, the noisy sounds of broadcasting in the environment, etc. In other words, the determination unit 122 may be used to determine the sound source direction of the above-mentioned fluctuating noise signal, for example, the direction of a passer-by (group) or a broadcasting speaker.

[0056] In some embodiments, if the number of microphones is at least two, the determination unit 122 may include a first calculation module and a second calculation module, respectively. The first calculation module may be used to calculate the time difference between the second type of signal components appearing in sequence in the acquisition sequence of each microphone. The second calculation module may be used to calculate the sound source direction of the second type of signal component based on the time difference. Since the time sequence of each microphone receiving the same sound signal is related to the direction of arrival of the sound source, based on this, the time difference between each microphone receiving the same sound signal is solved to determine the direction of the sound source.

[0057] As an implementation method, the time difference between each microphone receiving the same sound signal can be calculated based on the cross-correlation function between the sound signals collected by each microphone. The embodiment of the present application does not limit the specific implementation of the cross-correlation function.

[0058] In step S320, the classification unit 124 performs binary classification processing on the first sound signal according to the sound source direction of the second type of signal component to obtain two signal sets.

[0059] After the determination unit 122 determines the sound source direction of the second type of signal component, the classification unit 124 can divide the first sound signal into two signal sets, such as a first signal set and a second signal set, according to the sound source direction. As an example, the first signal set may include the first type of signal component, and the second signal set may include the second type of signal component. As another example, the first signal set may include the second type of signal component, and the second signal set may include the first type of signal component.

[0060] The embodiment of the present application does not limit the specific standard for the classification unit 124 to classify the first sound signal. For example, the first-class signal component in each signal point in the first sound signal can be estimated according to the probability that the second-class signal component exists in each signal point in the first sound signal. This will be described in detail later in conjunction with the accompanying drawings and will not be repeated here.

[0061] In step S330, the filtering unit 126 performs a fidelity processing on the signal set to which the second type of signal components belong, and performs a distortion processing on the signal set to which the first type of signal components belong, to obtain a second sound signal.

[0062] The embodiment of the present application does not limit the specific implementation method of the filtering unit 126 performing distortion processing on the signal set to which the first type of signal component belongs. As an implementation method, the filtering unit 126 can perform amplitude zeroing processing on the signal set to which the first type of signal component belongs, so that the first type of signal component is distorted. As another implementation method, the filtering unit 126 can perform phase shift processing on the signal set to which the first type of signal component belongs, so that the signal set to which the first type of signal component belongs and the signal set to which the second type of signal component belongs do not have an intersection in the time domain, thereby facilitating the use of any filtering method to filter the signal set to which the first type of signal component belongs, so that the first type of signal component is distorted.

[0063] The filtering unit 126 in the circuit preserves the second type of signal components and filters out the first type of signal components, that is, the wind noise signal collected by the microphone 110 is cut off at the spatial domain filter 120, and will not enter the noise reduction filter 130 to generate an inverted wind noise signal, so the headphone user will no longer hear obvious wind noise; the second type of signal components in the environment collected by the microphone 110 enter the noise reduction filter 130 through the spatial domain filter 120, and generate a corresponding noise reduction signal to offset the noise entering the user's ear canal, so that the headphones provided in the embodiment of the present application realize an active noise reduction mode that can resist wind noise compared to the existing active noise reduction mode.

[0064] When the earphone uses the second noise reduction mode, the spatial filter 120 performs a second filtering process on the first sound signal, which may include: Figure 4 Steps shown.

[0065] Specifically, in step S410 , the determination unit 122 determines the sound source direction of the second type signal component.

[0066] As mentioned above, when the second noise reduction mode is the selective transparency mode, the second type of signal component may include, for example, the voice of a companion speaking, a whistle from a certain direction of the user, and other directional sounds of interest to the user. That is, the determination unit 122 may be used to determine the sound source direction of the above-mentioned fluctuating noise signal, for example, the direction of a companion or a car.

[0067] Regarding the specific implementation method of the determination unit 122 determining the sound source direction of the second type of signal component, reference may be made to the above step S310, which will not be described again here.

[0068] In step S420, the classification unit 124 performs binary classification processing on the first sound signal according to the sound source direction of the second type of signal component to obtain two signal sets.

[0069] For the specific implementation of step S420, please refer to the description of step S320 above.

[0070] In step S430, the filtering unit 126 performs a fidelity processing on the signal set to which the first type of signal components belong, and performs a distortion processing on the signal set to which the second type of signal components belong, to obtain a second sound signal.

[0071] The embodiment of the present application does not limit the specific implementation method of the filtering unit 126 performing distortion processing on the signal set to which the second type of signal component belongs. For specific exemplary implementation methods, please refer to the related introduction of the filtering unit 126 performing distortion processing on the first type of signal component in the previous text, which will not be repeated here.

[0072] The filtering unit 126 in the circuit preserves the first type of signal components and filters out the second type of signal components, that is, the directional signal collected by the microphone 110 is cut off at the spatial filter 120, and will not enter the noise reduction filter 130 to generate an inverted directional signal. Therefore, valuable sounds can be heard by the headphone user through the physical space and will not be actively reduced; the first type of signal components in the environment collected by the microphone 110 enter the noise reduction filter 130 through the spatial filter 120, and generate a corresponding noise reduction signal to offset the worthless sounds entering the user's ear canal, so that the headphones provided in the embodiment of the present application realize a transparent mode that can selectively actively reduce noise compared to the existing transparent mode.

[0073] The active noise reduction mode of the headphones provided in the embodiment of the present application has the effect of resisting wind noise (effectively blocking the wind noise that the original active noise reduction system was forced to introduce into the human ear), and the transparency mode has the effect of actively suppressing worthless sounds (no longer distorting the ambient noise that the original transparency system was forced to compensate for). The active noise reduction function and transparency function of the existing headphones are optimized respectively, and the two optimizations are achieved under the same filtering system. To a certain extent, the underlying architecture of the two "opposite functions" of noise reduction and transparency are organically combined, and the joint design of the two algorithms significantly reduces the development difficulty and software cost.

[0074] As described above, the classification unit 124 can estimate the first type signal component in each signal point according to the probability that the second type signal component exists in each signal point in the first sound signal (equivalent to the proximity of the sound source direction of each signal point close to the second type signal component). Figure 5 , which is described in detail.

[0075] Figure 5 This is a schematic diagram of the structure of the classification unit 124 provided in one embodiment of the present application. Figure 5 As shown, the classification unit 124 may include a construction module 1242 , a convolution module 1244 , an estimation module 1246 , and a cropping module 1248 .

[0076] The construction module 1242 can be used to construct multiple basis function signals pointing to the sound source direction of the second type of signal component. It should be understood that the multiple basis function signals are linearly independent from each other and together construct a linear space.

[0077] The convolution module 1244 is connected to the construction module 1242, and can be used to perform convolution calculations on the first sound signal with the multiple basis functions constructed by the construction module 1242, and determine the probability of the existence of the second type of signal component in each signal point in the first sound signal based on the convolution result.

[0078] The estimation module 1246 is connected to the convolution module 1244 , and can be used to estimate the first type of signal component in each signal point according to the probability determined by the convolution module 1244 .

[0079] The embodiment of the present application does not limit the specific implementation method of the estimation module 1246 estimating the first type of signal components in each signal point according to probability. Exemplarily, the estimation module 1246 can identify signal points with a probability less than a certain threshold (for example, 50%, 80%, etc.) as pure random points (or pure points, which can be considered to contain only the first type of signal components and no second type of signal components). The so-called "pure random" corresponds to the characteristics of the first type of signal components (whether it is a pneumatic noise signal or an undirected fluctuation noise signal) that cannot be described by a deterministic function and the statistical characteristics are long-term stable. On the contrary, the directional fluctuation noise signals (second type of signal components) such as the aforementioned human voice, broadcast sound, whistle sound, etc. cannot be described by a deterministic function, but their statistical characteristics are long-term non-stationary. In addition, the estimation module 1246 can also count all pure random points to estimate the first type of signal components in each signal point, which is precisely the long-term stable characteristics of the statistical characteristics of the first type of signal components.

[0080] The clipping module 1248 can be used to clip each signal point into corresponding first-type signal components and second-type signal components to obtain two signal sets as described above, so that the subsequent filtering unit 126 performs corresponding processing according to the signal set to obtain the second sound signal.

[0081] The classification unit 124 performs binary classification processing on the first sound signal in the form of convolution calculation, which can save calculation cost.

[0082] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A headset, characterized in that: include: A microphone, used to collect a first sound signal, wherein the first sound signal includes a first type of signal component and a second type of signal component; A spatial domain filter connected to the microphone, and configured to filter the first sound signal to obtain a second sound signal; a noise reduction filter connected to the spatial domain filter, configured to receive the second sound signal and determine a noise reduction signal corresponding to the second sound signal; as well as A speaker, connected to the noise reduction filter, for playing the noise reduction signal; When the headset uses the first noise reduction mode, the spatial domain filter is used to perform a first filtering process on the first sound signal, so that the second type of signal components pass through and the first type of signal components are cut off, and the first noise reduction mode is an active noise reduction mode for resisting wind noise; wherein the first type of signal components include a pneumatic noise signal, and the second type of signal components include a directional wave noise signal; When the earphones use the second noise reduction mode, the spatial domain filter is used to perform a second filtering process on the first sound signal, so that the first type of signal components pass through and the second type of signal components are cut off, and the second noise reduction mode is a selective transparency mode; wherein the first type of signal components include omnidirectional fluctuating noise signals, and the second type of signal components include directional fluctuating sound signals.

2. The earphone according to claim 1, characterized in that The spatial domain filter includes a determination unit, a classification unit and a filtering unit, wherein the determination unit is connected to the microphone, the classification unit is connected to the determination unit, and the filtering unit is connected to the classification unit. The spatial domain filter is used to perform a first filtering process on the first sound signal, including: The determining unit is used to determine the sound source direction of the second type of signal component; The classification unit is used to perform binary classification processing on the first sound signal according to the sound source direction of the second type of signal component to obtain two signal sets; The filtering unit is used to perform fidelity processing on the signal set to which the second type of signal components belong, and perform distortion processing on the signal set to which the first type of signal components belong, so as to obtain the second sound signal.

3. The earphone according to claim 2, characterized in that The filtering unit performs distortion processing on the signal set to which the first type of signal component belongs, including: the filtering unit performs amplitude zeroing processing on the signal set to which the first type of signal component belongs; or, the filtering unit performs phase shift processing on the signal set to which the first type of signal component belongs, so that the signal set to which the first type of signal component belongs and the signal set to which the second type of signal component belongs have no intersection in the time domain.

4. The earphone according to claim 1, characterized in that The spatial domain filter includes a determination unit, a classification unit and a filtering unit, wherein the determination unit is connected to the microphone, the classification unit is connected to the determination unit, and the filtering unit is connected to the classification unit. The spatial domain filter is used to perform a second filtering process on the first sound signal, including: The determining unit is used to determine the sound source direction of the second type of signal component; The classification unit is used to perform binary classification processing on the first sound signal according to the sound source direction of the second type of signal component to obtain two signal sets; The filtering unit is used to perform fidelity processing on the signal set to which the first type of signal components belong, and perform distortion processing on the signal set to which the second type of signal components belong, so as to obtain the second sound signal.

5. The earphone according to claim 4, characterized in that: The filtering unit performs distortion processing on the signal set to which the second-category signal component belongs, including: the filtering unit performs amplitude zeroing processing on the signal set to which the second-category signal component belongs; or, the filtering unit performs phase shift processing on the signal set to which the second-category signal component belongs, so that the signal set to which the second-category signal component belongs and the signal set to which the first-category signal component belongs have no intersection in the time domain.

6. The earphone according to claim 2 or 4, characterized in that: The taxonomic units include: A construction module, used to construct a plurality of basis function signals pointing to the sound source direction of the second type of signal component; wherein the plurality of basis function signals are linearly independent from each other; a convolution module, configured to convolve the first sound signal with the plurality of basis function signals respectively, and determine, based on the convolution result, a probability that the second type of signal component exists in each signal point in the first sound signal; an estimation module, configured to estimate the first type of signal component in each signal point according to the probability; The cropping module is used to crop each signal point into the corresponding first-category signal component and the second-category signal component to obtain the two signal sets.

7. The earphone according to claim 6, characterized in that The estimation module is used to estimate the first type of signal components in each signal point according to the probability, including: the estimation module is used to identify the signal points whose probability is less than a threshold as purely random points, and count all the purely random points to estimate the first type of signal components in each signal point.

8. The earphone according to claim 2 or 4, characterized in that: The number of the microphones is at least two, and the determining unit includes: A first calculation module, used for calculating the time difference of the second type of signal components appearing successively in the acquisition sequence of each microphone; The second calculation module is used to calculate the sound source direction of the second type of signal component according to the time difference.

Images