WO2024216532A1 - Filtering method, filter apparatus, chip, and earphone - Google Patents

Abstract

The present application provides a filtering method, a filter apparatus, a chip, and an earphone, which are capable of improving the transparency effect of the earphone. The earphone comprises an extra-aural microphone, sound signals collected by the extra-aural microphone comprising signals of a first frequency band and signals of a second frequency band, and the first frequency band being lower than the second frequency band. The method comprises: processing a sound signal on the basis of a first transparent filter so as to obtain a first filtered signal of the sound signal, wherein the first transparent filter causes a greater gain in a signal of a first frequency band than in a signal of a second frequency band; performing noise reduction processing on the sound signal, and on the basis of a second transparent filter, processing the noise-reduced sound signal so as to obtain a second filtered signal of the sound signal, wherein the second transparent filter causes a greater gain in a signal of the second frequency band than in a signal of the first frequency band; and according to the first filtered signal and the second filtered signal, determining a compensation signal for the earphone.

Description

Filtering method, filtering device, chip and earphone Technical Field

Embodiments of the present application relate to the field of audio processing, and more specifically, to a filtering method, a filtering device, a chip, and a headset.

Background Art

Headphones, especially in-ear headphones, block most of the ambient sound when worn normally. Usually, headphone wearers want to hear the ambient sound clearly in certain scenarios, so the headphones need to have a transparent mode. The transparent mode of the headphones can use a transparent filter to lift the sound outside the headphones, thereby restoring the external sound inside the ear. The transparent effect of the headphones directly affects the user's auditory experience. Therefore, how to improve the transparent effect of the headphones has become a problem that needs to be solved.

Summary of the invention

The embodiments of the present application provide a filtering method, a filtering device, a chip and headphones, which can improve the transparency effect of the headphones.

In a first aspect, a filtering method is provided, which is applied to headphones with a transparent mode, the headphones comprising an external ear microphone, the sound signal collected by the external ear microphone comprising a signal of a first frequency band and a signal of a second frequency band, at least part of the frequency band in the first frequency band is lower than at least part of the frequency band in the second frequency band, the method comprising: processing the sound signal based on a first transparent filter to obtain a first filtered signal of the sound signal, the first transparent filter being a hardware transparent filter, and the gain of the first transparent filter for the signal of the first frequency band in the sound signal being greater than the gain of the first transparent filter for the signal of the second frequency band in the sound signal; performing background noise reduction processing on the sound signal, and processing the sound signal after background noise reduction based on a second transparent filter to obtain a second filtered signal of the sound signal, the second transparent filter being a software transparent filter, and the gain of the second transparent filter for the signal of the second frequency band in the sound signal after background noise reduction being greater than the gain of the second transparent filter for the signal of the first frequency band in the sound signal after background noise reduction; and determining a compensation signal of the headphone according to the first filtered signal and the second filtered signal.

The software transparent filter includes, for example, an FIR filter or an IIR filter, and the hardware transparent filter includes, for example, an FIR filter or an IIR filter.

In one implementation, the performing background noise reduction processing on the sound signal includes: acquiring a pre-estimated background noise signal; and determining the sound signal after background noise reduction according to the background noise signal and the sound signal.

In one implementation, the headset also includes a speaker and an in-ear microphone, and the method also includes: determining a secondary transfer function based on an audio signal output by the speaker and an in-ear signal collected by the in-ear microphone; determining the first transparent filter based on an external ear signal collected by the external ear microphone when the headset is worn and a signal received in the ear when the headset is not worn.

In one implementation, the method also includes: determining energy values of multiple frequency bands of the sound signal; and adjusting the energy values of the multiple frequency bands of the compensation signal according to the energy values of the multiple frequency bands of the sound signal, so that the energy values of the multiple frequency bands of the compensation signal are less than or equal to energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the energy values of the multiple frequency bands of the compensation signal are adjusted separately according to the energy values of the multiple frequency bands of the sound signal, including: determining the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band; and adjusting the energy value of each frequency band of the compensation signal according to the adjustment amount corresponding to each frequency band.

In one implementation, determining the adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band includes: determining the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the method also includes: determining energy values of multiple frequency bands of the sound signal; the first transparent filter is also used to adjust the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the first filtered signal are less than or equal to energy thresholds corresponding to each of the multiple frequency bands; the second transparent filter is also used to adjust the energy values of the multiple frequency bands of the sound signal after background noise reduction so that the energy values of the multiple frequency bands of the second filtered signal are less than or equal to energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the first transparent filter is specifically configured to, according to the energy value of each frequency band of the plurality of frequency bands of the sound signal and the energy threshold corresponding to each frequency band, Determine the adjustment amount corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal according to the adjustment amount corresponding to each frequency band; the second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal after background noise reduction according to the adjustment amount corresponding to each frequency band.

In one implementation, the first transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band; the second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the first frequency band is less than or equal to a preset first frequency threshold, the second frequency band is greater than or equal to a preset second frequency threshold, and the first frequency threshold and the second frequency threshold are between 2 kHz and 3 kHz.

In a second aspect, a filtering device is provided, which is applied to headphones with a transparent mode, the headphones comprising an external ear microphone, the sound signal collected by the external ear microphone comprising a signal of a first frequency band and a signal of a second frequency band, at least part of the frequency band in the first frequency band is lower than at least part of the frequency band in the second frequency band, the filtering device comprising: a first transparent filter, used to process the sound signal to obtain a first filtered signal of the sound signal, the first transparent filter is a hardware transparent filter, and the gain of the first transparent filter on the signal of the first frequency band in the sound signal is greater than the gain of the first transparent filter on the signal of the second frequency band in the sound signal; a processing module, used to perform background noise reduction processing on the sound signal; a second transparent filter, used to process the sound signal after background noise reduction to obtain a second filtered signal of the sound signal, the second transparent filter is a software transparent filter, and the gain of the second transparent filter on the signal of the second frequency band in the sound signal after background noise reduction is greater than the gain of the second transparent filter on the signal of the first frequency band in the sound signal after background noise reduction;

The processing module is further configured to determine a compensation signal of the earphone according to the first filtered signal and the second filtered signal.

The software transparent filter includes, for example, an FIR filter or an IIR filter. The transmission filter includes, for example, a FIR filter or an IIR filter.

In one implementation, the processing module is specifically used to: obtain a pre-estimated background noise signal; and determine the sound signal after background noise reduction according to the background noise signal and the sound signal.

In one implementation, the earphone also includes a speaker and an in-ear microphone, and the processing module is further used to: determine the secondary transfer function based on the audio signal output by the speaker and the in-ear signal collected by the in-ear microphone; determine the first transparent filter based on the external ear signal collected by the external ear microphone when the earphone is worn and the received signal in the ear when the earphone is not worn.

In one implementation, the processing module is also used to: determine the energy values of multiple frequency bands of the sound signal; and adjust the energy values of the multiple frequency bands of the compensation signal according to the energy values of the multiple frequency bands of the sound signal, so that the energy values of the multiple frequency bands of the compensation signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the processing module is specifically used to: determine the adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band; and adjust the energy value of each frequency band of the compensation signal based on the adjustment amount corresponding to each frequency band.

In one implementation, the processing module is specifically configured to determine the adjustment amount corresponding to each frequency band according to a difference between the energy value of each frequency band and an energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the first transparent filter is further used to adjust the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the first filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands; the second transparent filter is further used to adjust the energy values of the multiple frequency bands of the sound signal after background noise reduction so that the energy values of the multiple frequency bands of the second filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the first transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal according to the adjustment amount corresponding to each frequency band; the second transparent filter is specifically used to According to the energy value of each frequency band in the multiple frequency bands of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band, an adjustment amount corresponding to each frequency band is determined, and according to the adjustment amount corresponding to each frequency band, the energy value of each frequency band of the sound signal after background noise reduction is adjusted.

In one implementation, the first transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band; the second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the first frequency band is less than or equal to a preset first frequency threshold, the second frequency band is greater than or equal to a preset second frequency threshold, and the first frequency threshold and the second frequency threshold are between 2 kHz and 3 kHz.

According to a third aspect, a chip is provided, comprising a processor and a memory, wherein the memory stores computer instructions, and the processor calls the computer instructions to enable the device to implement the filtering method described in the first aspect or any implementation method of the first aspect.

In a fourth aspect, an earphone is provided with a transparency mode, the earphone comprising: a speaker; an external microphone; an internal microphone; and the filtering device described in the second aspect or any implementation of the second aspect, or the chip described in the third aspect.

In a fifth aspect, a filtering method is provided, which is applied to headphones with a transparency mode, the headphones including an external ear microphone, and the method comprising: determining energy values of multiple frequency bands of sound signals collected by the external ear microphone; adjusting the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands are less than or equal to energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the adjusting the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands includes: processing the sound signal based on a transparent filter to obtain a filtered signal; wherein, the adjusting the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands includes: adjusting the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the adjusting the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to energy thresholds corresponding to each of the multiple frequency bands includes: based on the transparent filter or another filter other than the transparent filter, adjusting the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the adjusting the energy values of the multiple frequency bands of the filtered signal includes: determining an adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and an energy threshold corresponding to each frequency band, and adjusting the energy value of each frequency band of the sound signal based on the adjustment amount corresponding to each frequency band.

In one implementation, determining the adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band includes: determining the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the first frequency band is less than a preset first frequency threshold, the second frequency band is greater than or equal to a preset second frequency threshold, and the first frequency threshold and the second frequency threshold are between 2 kHz and 3 kHz.

In a sixth aspect, a filtering device is provided, which is applied to headphones with a transparent mode, wherein the headphones include an external ear microphone for collecting sound signals. The filtering device includes: a processing module for determining energy values of multiple frequency bands of the sound signal; and a filter module for adjusting the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands are less than or equal to energy thresholds corresponding to the multiple frequency bands.

In one implementation, the filter module includes a transparent filter, which is used to: process the sound signal to obtain a filtered signal; wherein the transparent filter is also used to: adjust the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to the energy thresholds corresponding to the multiple frequency bands; or, the filter module further includes another filter other than the transparent filter, which is used to: receive the filtered signal and adjust the energy values of the multiple frequency bands of the filtered signal. The values are adjusted so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to the energy thresholds corresponding to the multiple frequency bands respectively.

In one implementation, the transparent filter or the other filter is specifically used to: determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the compensation signal according to the adjustment amount corresponding to each frequency band.

In one implementation, the transparent filter or the another filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among N frequency bands of the sound signal, where N is a positive integer.

In the seventh aspect, a chip is provided, comprising a processor and a memory, wherein the memory stores computer instructions, and the processor calls the computer instructions to enable the device to implement the filtering method described in the fifth aspect or any implementation method of the fifth aspect.

In an eighth aspect, an earphone is provided with a transparency mode, the earphone comprising: a speaker; an external microphone; an internal microphone; and the filtering device described in the sixth aspect or any implementation of the sixth aspect, or the chip described in the seventh aspect.

Based on the above technical solution, the first transparent filter and the second transparent filter are used to process the sound signal collected by the external microphone respectively. The first transparent filter is a hardware transparent filter, and the second transparent filter is a software transparent filter. Usually, the passive isolation of the earphone to the high-frequency signal is strong, and a higher gain needs to be added to the high-frequency signal to lift it, resulting in the background noise being mainly concentrated in the high-frequency part. Therefore, the gain of the hardware transparent filter to the low- and medium-frequency signals in the sound signal is greater than the gain to the high-frequency signal, and the first filtered signal obtained is mainly composed of low- and medium-frequency signals; in addition, the sound signal is subjected to background noise reduction processing, and the gain of the software transparent filter to the high-frequency signal in the sound signal after background noise reduction is greater than the gain to the low- and medium-frequency signals. In this way, the high-frequency signal after background noise reduction is mainly obtained in the second filtered signal, which can be used to compensate for the high-frequency signal in the first filtered signal. The compensation signal of the earphone can be obtained by combining the first filtered signal and the second filtered signal. Since the high-frequency signal in the second filtered signal is a signal after background noise reduction processing, the problem of large background noise in the compensation signal is solved. In addition, since the sound that the human ear hears is the result of the superposition of the sound processed by the transparent filter and the sound leaking to the human ear, when the wearer speaks to himself, the frequency of the human voice is concentrated in the middle and low frequencies. The software transparent filter mainly processes the high-frequency signal and suppresses the middle and low-frequency signal. The signal can greatly reduce the processing delay of the software transparent filter on the sound signal, reduce the time difference between the processed sound and the sound leaked to the human ear, and reduce the sound distortion heard by the wearer.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic flow chart of a filtering method according to an embodiment of the present application.

FIG. 2 is a schematic diagram of an earphone according to an embodiment of the present application.

FIG. 3 is a schematic diagram of a possible implementation of the method shown in FIG. 1 .

FIG. 4 is a schematic diagram of a first transparent filter in an earphone.

FIG. 5 is a schematic diagram of another possible implementation of the method shown in FIG. 1 .

FIG. 6 is a schematic diagram of another possible implementation of the method shown in FIG. 1 .

FIG. 7 is a schematic diagram of another possible implementation of the method shown in FIG. 1 .

FIG. 8 is a schematic diagram of a possible specific implementation of the method shown in FIG. 1 .

FIG. 9 is a schematic flow chart of a filtering method according to another embodiment of the present application.

FIG. 10 is a schematic block diagram of a filtering device according to an embodiment of the present application.

FIG. 11 is a schematic block diagram of a filtering device according to another embodiment of the present application.

DETAILED DESCRIPTION

The technical solution in this application will be described below in conjunction with the accompanying drawings.

The headphone's transparent mode can use the Hear Through (HT) filter to lift the sound outside the headphone, thereby restoring the external sound inside the ear. However, this will also amplify the background noise of the headphone's internal circuit, causing the headphone's playback sound to carry a large background noise. In a quiet environment such as an office, the microphone needs to have a very small transparent sound. At this time, a large part of the input signal is the background noise of the circuit. The input signal is amplified by the HT filter, making the sound of the background noise particularly obvious. Reducing the background noise of the circuit itself through a hardware circuit requires a higher cost, and it is also difficult to eliminate the background noise.

Noise is also one of the important factors that restrict the bandwidth of transparency. Considering the noise floor, the bandwidth of the headphones currently on the market is usually around 2KHz-3.5KHz. To ensure the transparency effect within this frequency range, if the noise floor problem can be solved, a larger bandwidth can be achieved when designing the HT filter, and it can also be applied to hearing aids and other devices to solve the noise floor problem, which is less harmful to the human ear.

Usually, software transparency can be used to remove the circuit noise mixed in the input signal and reduce the noise to below the sound threshold that the human ear can hear. It can solve the problem of background noise to a certain extent, but software transparency has a greater delay than hardware transparency, which will cause the wearer of the headset to hear his or her own voice distorted when speaking. This is because there is a time difference between the compensating human voice emitted by the speaker of the headset and the human voice transmitted by bone conduction.

To this end, an embodiment of the present application provides a filtering method applied to headphones with a transparency mode, which processes the collected sound signals by combining software transparency and hardware transparency, thereby solving the above-mentioned background noise problem and sound distortion problem to a certain extent and improving the transparency effect of the headphones.

FIG. 1 is a schematic flow chart of a filtering method according to an embodiment of the present application. The method 100 shown in FIG. 1 is applied to headphones with a transparent mode, for example, the headphones 10 shown in FIG. 2. The headphones described in the embodiment of the present application should be understood in a broad sense, and include not only headphones in the traditional sense shown in FIG. 2, but also other types of sound players such as auxiliary or hearing aid devices. As shown in FIG. 2, the headphones 10 include a speaker 11, an external microphone 12, and an internal microphone 13. The external microphone 12 can be called a feed forward (FF) microphone or a reference microphone, which is used to collect sound signals transmitted from outside the ear, or external ear data. The internal microphone 13 can be called a feedback (FB) microphone, which is used to collect sound signals in the ear, or internal ear data. In the transparent mode, the sound signal collected by the external microphone 12 is processed by the transparent filter to output a compensation signal, and the residual signal leaking from the outside of the headphones 10 into the human ear is superimposed, so that the human ear can hear the sound of the complete external environment. When the user wears the earphone 10 to talk with other people, the user does not need to take off the earphone 10, but can directly switch the earphone 10 to the transparent mode to achieve a clear conversation with the other party.

The method 100 shown in FIG1 can be executed by a filtering device in the headset 10, for example, by a chip in the headset 10, such as an audio codec chip or a Bluetooth audio system chip (System on a Chip, SOC). The sound signal collected by the external microphone 12 includes a signal of a first frequency band and a signal of a second frequency band. Among them, at least part of the frequency band in the first frequency band is lower than at least part of the frequency band in the second frequency band. Here, the first frequency band can be regarded as a medium and low frequency signal, and the second frequency band can be regarded as a high frequency signal. Optionally, the first frequency band is less than or equal to a preset first frequency threshold, and the second frequency band is greater than or equal to a preset second frequency threshold. The first frequency threshold is, for example, between 2KHz and 3KHz, and the second frequency threshold is, for example, between 2KHz and 3KHz.

The first frequency threshold and the second frequency threshold may be equal, for example, the first frequency threshold and the second frequency threshold are both 2.5KHz, or the first frequency threshold and the second frequency threshold are both 2.8KHz. The first frequency threshold and the second frequency threshold may also be unequal, for example, the first frequency threshold is greater than the second frequency threshold, such as the first frequency threshold is 2.8KHz and the second frequency threshold is 2.5KHz; for another example, The first frequency threshold is smaller than the second frequency threshold. For example, the first frequency threshold is 2.5 KHz, and the second frequency threshold is 2.8 KHz.

As shown in FIG. 1 , method 100 includes part or all of the following steps.

In step 110, the sound signal collected by the external ear microphone 12 is processed based on the first transparent filter to obtain a first filtered signal of the sound signal.

The first transparent filter is a hardware transparent filter, also known as a hardware transparent module, which can amplify the gain of signals in a specific frequency band. The hardware transparent filter is, for example, a hardware filter circuit. In the embodiment of the present application, the first transparent filter is used to suppress the signal in the second frequency band of the collected sound signal, that is, to transparently transmit the signal in the first frequency band, and the obtained first filtered signal is mainly composed of the signal in the first frequency band. In other words, the gain of the first transparent filter on the signal in the first frequency band of the sound signal is greater than the gain of the first transparent filter on the signal in the second frequency band of the sound signal.

The gain of the first transparent filter for the signal of the first frequency band can be greater than 1, equal to 1, or less than 1. When the gain is greater than 1, it indicates that the signal of the first frequency band is amplified, when the gain is less than 1, it indicates that the signal of the first frequency band is reduced, and when the gain is equal to 1, it indicates that the signal of the first frequency band is not amplified or reduced. Similarly, the gain of the first transparent filter for the signal of the second frequency band can be greater than 1, equal to 1, or less than 1. When the gain is greater than 1, it indicates that the signal of the second frequency band is amplified, when the gain is less than 1, it indicates that the signal of the second frequency band is reduced, and when the gain is equal to 1, it indicates that the signal of the second frequency band is not amplified or reduced. For example, the gain of the first transparent filter for the signal of the first frequency band and the gain of the first transparent filter for the signal of the second frequency band are both greater than 1 or less than 1; or, the gain of the first transparent filter for the signal of the first frequency band is greater than 1, and the gain of the first transparent filter for the signal of the second frequency band is less than or equal to 1; or, the gain of the first transparent filter for the signal of the first frequency band is greater than or equal to 1, and the gain of the first transparent filter for the signal of the second frequency band is less than 1.

In step 120, the sound signal collected by the external ear microphone 12 is subjected to background noise reduction processing, and the sound signal after background noise reduction is processed based on a second transparent filter to obtain a second filtered signal of the sound signal.

The second transparent filter is a software transparent filter, also known as a software transparent module. The software transparent filter is a digital filter designed in the program, which can amplify the gain of the signal in a specific frequency band through a filtering algorithm. For example, the software transparent filter can include a finite impulse response (FIR) filter or an infinite impulse response (IIR) filter to perform software filtering on the sound signal after the background noise is eliminated; for another example, the hardware transparent filter can include a finite impulse response (FIR) filter or an infinite impulse response (IIR) filter. The sound signal collected by the external ear microphone 12 is filtered by hardware by means of an infinite impulse response (IIR) filter or the like. In the embodiment of the present application, the second transparent filter is used to suppress the signal of the first frequency band in the collected sound signal, that is, to transparently filter the signal of the second frequency band, and the obtained second filtered signal is mainly composed of the signal of the second frequency band. In other words, the gain of the second transparent filter on the signal of the second frequency band in the sound signal is greater than the gain of the second transparent filter on the signal of the first frequency band in the sound signal.

The gain of the second transparent filter for the signal of the first frequency band can be greater than 1, equal to 1, or less than 1. When the gain is greater than 1, it indicates that the signal of the first frequency band is amplified, when the gain is less than 1, it indicates that the signal of the first frequency band is reduced, and when the gain is equal to 1, it indicates that the signal of the first frequency band is not amplified or reduced. Similarly, the gain of the second transparent filter for the signal of the second frequency band can be greater than 1, equal to 1, or less than 1. When the gain is greater than 1, it indicates that the signal of the second frequency band is amplified, when the gain is less than 1, it indicates that the signal of the second frequency band is reduced, and when the gain is equal to 1, it indicates that the signal of the second frequency band is not amplified or reduced. For example, the gain of the second transparent filter for the signal of the first frequency band and the gain of the second transparent filter for the signal of the second frequency band are both greater than 1 or less than 1; or, the gain of the second transparent filter for the signal of the first frequency band is less than or equal to 1, and the gain of the second transparent filter for the signal of the second frequency band is greater than 1; or, the gain of the second transparent filter for the signal of the first frequency band is less than 1, and the gain of the second transparent filter for the signal of the second frequency band is greater than 1.

In step 130, a compensation signal of the earphone 10 is determined according to the first filtered signal output by the first transparent filter and the second filtered signal output by the second transparent filter.

The compensation signal is used to compensate for the signal attenuated by the passive isolation and is output by the loudspeaker 11. The compensation signal is superimposed on the residual signal leaked to the human ear to form the sound finally heard by the human ear.

In the embodiment of the present application, a first transparent filter and a second transparent filter are used to process the sound signal collected by the external microphone 12 respectively. The first transparent filter is a hardware transparent filter, and the second transparent filter is a software transparent filter. Usually, the passive isolation of the earphone 10 to the high-frequency signal is strong, and a higher gain needs to be added to the high-frequency signal to lift it, resulting in the background noise being mainly concentrated in the high-frequency part. Therefore, the hardware transparent filter has a greater gain for the mid- and low-frequency signals in the sound signal than for the high-frequency signals, and the obtained first filtered signal is mainly composed of mid- and low-frequency signals. In addition, the sound signal is subjected to background noise reduction processing, and the software transparent filter has a greater gain for the high-frequency signals in the sound signal after background noise reduction than for the mid- and low-frequency signals. In this way, the obtained second filtered signal is mainly composed of the high-frequency signals after background noise reduction, which can be used to compensate for the high-frequency signals in the first filtered signal. The first filtered signal and the second filtered signal are combined to obtain a compensation signal for the earphone. Since the second The high-frequency signal in the filtered signal is the signal that has been processed by the noise reduction process, thus solving the problem of large noise in the compensation signal. In addition, since the sound that the human ear finally hears is the superposition of the sound processed by the transparent filter and the sound leaked to the human ear, when the wearer speaks to himself, the frequency of the human voice is concentrated in the medium and low frequencies. The software transparent filter mainly processes the high-frequency signal and suppresses the medium and low frequency signals, which can greatly reduce the processing delay of the software transparent filter on the sound signal, reduce the time difference between the processed sound and the sound leaked to the human ear, and reduce the sound distortion heard by the wearer.

It can be understood that the background noise signal is usually the background noise introduced by various circuits set between the external ear microphone 12 and the transparent filter, such as the analog to digital converter (ADC) circuit. Usually, the headset 10 has strong passive isolation for high-frequency signals. Whether a hardware transparent filter or a software transparent filter is used, a higher gain needs to be added to the high-frequency signal to increase it, resulting in the background noise being mainly concentrated in the high-frequency part. Since the sound signal collected by the external ear microphone 12 is subjected to background noise reduction processing in step 120, the background noise carried in the high-frequency signal in the second filtered signal output by the second transparent filter is significantly reduced.

In the embodiment of the present application, the sound signal collected by the external ear microphone 12 refers to the sound signal received from the external ear microphone 12 by the first transparent filter and the second transparent filter, that is, the sound signal carrying background noise.

In one implementation, in step 120, the sound signal is subjected to noise reduction processing, including: obtaining a pre-estimated noise floor signal; determining the sound signal after noise floor reduction based on the noise floor signal and the sound signal collected by the external ear microphone 12. The noise floor signal can be obtained, for example, by experimental testing or the like. By using the pre-estimated noise floor signal to offset the part of the noise floor signal carried in the sound signal, the actual signal outside the external ear microphone 12 can be obtained. For example, the noise floor signal has an opposite sign to that of the sound signal, and by adding the noise floor signal to the sound signal, the noise floor signal can be removed from the sound signal.

For example, as shown in FIG3 , in the hardware transparent stage, after the sound signal collected by the external ear microphone 12 passes through the first transparent filter, the high-frequency signal is suppressed, and a first filtered signal mainly composed of medium and low frequency signals is output; in the software transparent stage, the background noise estimation module outputs a pre-estimated background noise signal, and the sign of the background noise signal is negative. By adding the background noise signal to the sound signal, the part of the background noise signal carried in the sound signal can be eliminated to obtain a sound signal with reduced background noise. After the sound signal with reduced background noise passes through the second transparent filter, the medium and low frequency signals are suppressed, and a second filtered signal mainly composed of high frequency signals is output. By merging the first filtered signal and the second filtered signal, a compensation signal output by the speaker 11 can be obtained. Signal.

The background noise signal can be a fixed value; or, the background noise signal can be estimated in real time based on the sound signal currently collected by the external ear microphone 12. For example, the characteristics of the background noise signal are collected and counted, and a reference value of the background noise signal is preset. The reference value can be obtained by, for example, expanding the background noise signal of the circuit in the frequency domain, and the reference value is adjusted according to the sound signal currently collected by the external ear microphone 12 to obtain an estimated value of the background noise signal, thereby realizing real-time estimation of the background noise signal.

If only a hardware transparent filter is used, as mentioned above, since the background noise is mainly concentrated in the high-frequency part, when the sound signal is lifted, the background noise will be amplified at the same time; if only a software transparent filter is used, although the corresponding background noise can be estimated by software and the background noise can be filtered out before the speaker 11 outputs the compensation signal, the software transparency has a greater delay than the hardware transparency. Since the sound that the human ear finally hears is formed by the superposition of the compensation sound processed by the transparent filter and the residual sound leaked to the human ear, the residual sound is physically transmitted and its delay can be ignored. If there is a large delay between the compensation sound input to the human ear and the residual sound leaked to the human ear, the wearer will hear the distortion of his own voice when speaking to himself.

The embodiment of the present application adopts a combination of hardware transparency and software transparency. Since the background noise is mainly concentrated in the high-frequency part, the high-frequency signal in the sound signal is suppressed by a hardware transparency filter and the mid- and low-frequency signals are transparent. This can avoid introducing more background noise of high-frequency signals. The background noise in the sound signal is estimated (estimate noise floor), and the mid- and low-frequency signals in the sound signal after the background noise is reduced are suppressed by a software transparency filter and the high-frequency signal is transparent. This can avoid the software transparency filter spending more time on processing the mid- and low-frequency signals concentrated on the human voice, reducing the time difference between the compensated sound and the residual sound, or the time delay between the compensated sound and the residual sound, and reducing the sound distortion heard by the wearer.

Optionally, DSP hardware may be used for acceleration in the second transparent filter, and most of the code implementation may be written in assembly language to complete background noise filtering and transparent filtering in fewer instruction cycles, thereby achieving ultra-low latency.

The first filtered signal mainly composed of medium and low frequency signals output by the hardware transparent filter and the second filtered signal mainly composed of high frequency signals output by the software transparent filter can be combined to restore the signal of the entire frequency band. In this way, the hardware transparent filter and the software transparent filter can complement each other, reducing the background noise in the compensation signal and reducing the time difference between the compensation signal and the residual signal.

In one implementation, the method 100 further includes: determining the secondary transfer function according to the audio signal output by the speaker 11 and the in-ear signal collected by the in-ear microphone 13; ... The first transparent filter is determined by the external ear signal in the wearing state and the signal received in the ear when the earphone 10 is not worn. It can be understood that the signal received in the ear when the earphone 10 is not worn refers to the signal collected from the ear when the earphone 10 is not worn. For example, when the ear is not worn, an additional microphone can be used to collect the signal in the ear. When determining the secondary transfer function, the ear signal collected by the ear microphone 13 is usually the signal collected by the ear microphone 13 in the earphone 10 when the earphone 10 is worn.

The design of the first transparent filter is specifically described in conjunction with FIG4. As an example, as shown in FIG4, the channel from the speaker 11 to the in-ear microphone 13 is a physical channel, and its modeling is called a secondary transfer function, denoted as SP. The secondary transfer function is a transfer function representing the transfer function from the speaker 11 to the in-ear microphone 13, also known as a secondary path transfer function. The purpose of transparency is to allow the wearer of the headset 10 to hear the sound of the outside world without passive isolation of the headset 10. In order to achieve this goal, it is necessary to compensate for the signal attenuated due to passive isolation, and output the compensation signal through the speaker 11. As shown in FIG4, it is usually hoped that the signal at the in-ear microphone 13 is consistent with the signal at the external ear microphone 12, so that the signal played by the speaker 11 when the headset 10 is worn is superimposed with the residual signal in the ear after the secondary transmission path, and the signal received in the ear when the headset 10 is not worn. Here, the path between the external ear microphone 12 and the speaker 11 is denoted as HT. Specifically, for the same sound source, the signal collected in the ear when the earphone 10 is not worn is recorded as D, the size of the residual signal collected by the ear microphone 13 when the earphone 10 is worn is recorded as E, and the signal collected by the external ear microphone 12 is recorded as FF, ignoring the difference between the position of the ear microphone 13 and the position of signal collection when the earphone 10 is not worn. The optimization goal of the first transparent filter is FF*HT*SP+E=D, and the corresponding first transparent filter can be obtained in this way. In addition, the first transparent filter needs to be designed to suppress high-frequency signals, so that the frequency band amplified by the hardware is concentrated in the middle and low frequencies.

When the earphone 10 is in the transparent mode, although it is hoped to maintain a clear perception of the ambient sound, there are often loud noises in the environment, such as firecrackers, knocking sounds during construction, etc. The restoration of these sounds will cause sensory discomfort to the wearer, and too high a volume may exceed the limit of the audio system's expression, resulting in clipping problems. Imagine that when the wearer passes a construction site through the earphone voice navigation, he will be in a dilemma. In order to hear the ambient sound clearly, the wearer needs to turn on the transparent mode of the earphone. Turning on the transparent mode will restore all ambient sounds, while louder external sounds will cause ear discomfort. Turning off the transparent mode will isolate the earphone and effectively reduce the external noise from entering the eardrum, but it is not convenient for obtaining the ambient sound.

To this end, the present application also provides a solution for environmental self-adaptation and transparency, which can reduce the large The transparency of the volume band, while the sound details of other frequency bands can be restored in the ear to the greatest extent, which is especially important for headphones with good passive isolation effects.

In one implementation, method 100 also includes: obtaining energy values of multiple frequency bands of the sound signal collected by the external ear microphone 12; and adjusting the energy values of multiple frequency bands of the compensation signal according to the energy values of the multiple frequency bands of the sound signal, so that the energy values of the multiple frequency bands of the compensation signal are less than or equal to the energy thresholds corresponding to the multiple frequency bands.

In this embodiment, the sound signal collected by the external microphone 12 is divided into multiple frequency bands, or multiple frequency bands, and the energy value of each frequency band is counted. According to the energy value of each frequency band in the sound signal, the energy value of each frequency band in the above-mentioned compensation signal is adjusted respectively, so that the energy values of the multiple frequency bands of the compensation signal are less than or equal to their respective corresponding energy thresholds, thereby avoiding the external sound signal being too high to cause sensory discomfort to the wearer, and realizing adaptive processing of ambient sound.

The multiple frequency bands are, for example, frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, where N is a positive integer. For frequency bands whose energy values do not exceed their corresponding energy thresholds, no adjustment may be performed.

Here, the energy value may be, for example, an energy peak value or an energy mean value, and the following description is made by taking the energy peak value as an example. The energy thresholds corresponding to different frequency bands may be determined, for example, by experimental tests or response curves. The energy thresholds corresponding to any two frequency bands in the N frequency bands may be equal or unequal.

Optionally, the energy values of multiple frequency bands of the compensation signal are adjusted respectively according to the energy values of multiple frequency bands of the sound signal, including: determining the adjustment amount corresponding to each frequency band according to the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band; and adjusting the energy value of each frequency band of the compensation signal according to the adjustment amount corresponding to each frequency band.

For example, the adjustment amount corresponding to each frequency band may be determined according to the difference between the energy value of each frequency band in the multiple frequency bands and the energy threshold corresponding to each frequency band. The larger the difference, the larger the adjustment amount; the smaller the difference, the smaller the adjustment amount.

For example, as shown in FIG5 , after the first transparent filter outputs the first filter signal and the second transparent filter outputs the second filter signal, the first filter signal and the second filter signal are combined to obtain a corresponding compensation signal, and the compensation signal can also be subjected to the above-mentioned adaptive processing. Specifically, the time domain signal collected by the external ear microphone 12 can be expanded in the frequency domain, and the effective frequency band of the sample can be divided into N frequency bands by, for example, an octave, and the energy peaks of the N frequency bands are counted, and the energy of multiple frequency bands whose energy peaks exceed the corresponding energy thresholds is adjusted. In the process of detecting the energy peak value and estimating the adjustment amount (environment adaptive energy detection and suppression estimate), the adjustment amount corresponding to the frequency band can be calculated according to the difference between the energy peak value of each frequency band in multiple frequency bands and its corresponding energy threshold, so as to adjust the energy value of the frequency band according to the adjustment amount, so that the energy value of the frequency band is suppressed below the corresponding energy threshold. This process is also called subband suppression. In practical applications, the external ear microphone 12 collects sound signals in real time, and can adjust the energy values of each frequency band in the compensation signal in real time according to the energy values of each frequency band in the real-time collected sound signals.

For example, N frequency bands correspond to N energy thresholds respectively. Assuming that the energy peak values of M frequency bands among the N frequency bands of the sound signal exceed their corresponding energy thresholds, the energy values of the M frequency bands are adjusted according to the difference between the energy peak values of the M frequency bands of the sound signal and their corresponding energy thresholds. For example, the adjustment amount of the i-th frequency band among the M frequency bands can be equal to the difference between the energy peak value of the i-th frequency band and the energy threshold corresponding to the i-th frequency band, where i ranges from 1 to M, and the energy value of the i-th frequency band is adjusted according to the adjustment amount of the i-th frequency band.

In the embodiment of the present application, the two software parts of software transparent processing and adaptive processing of ambient sound can also be combined into one software serial implementation. For example, as shown in FIG6 , the process of detecting the energy peak and estimating the adjustment amount can be completed before the second transparent filter performs software transparent processing on the sound signal.

In another implementation, method 100 also includes: obtaining energy values of multiple frequency bands of the sound signal collected by the external ear microphone 12; the first transparent filter is also used to adjust the energy values of multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the first filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands; the second transparent filter is also used to adjust the energy values of multiple frequency bands of the sound signal after background noise reduction so that the energy values of the multiple frequency bands of the second filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In this embodiment, as shown in FIG7 , the first transparent filter can suppress the high-frequency signal in the sound signal and adjust the energy value of each frequency band in the sound signal, so that the output first filtered signal is mainly composed of medium and low frequency signals, and the energy of each frequency band in the first filtered signal is less than or equal to the corresponding energy threshold. Similarly, the second transparent filter can suppress the medium and low frequency signals in the sound signal after the background noise is reduced and adjust the energy value of each frequency band in the sound signal after the background noise is reduced, so that the output second filtered signal is mainly composed of high frequency signals, and the energy of each frequency band in the second filtered signal is less than or equal to the corresponding energy threshold.

Optionally, the first transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal according to the adjustment amount corresponding to each frequency band; the second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal after background noise reduction according to the adjustment amount corresponding to each frequency band.

For example, the first transparent filter can determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band; the second transparent filter can determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal after noise reduction and the energy threshold corresponding to each frequency band.

For the above-mentioned adaptive processing of ambient sound, for example, as shown in FIG5 or FIG6, after the first filtered signal and the second filtered signal are obtained respectively through the first transparent filter and the second transparent filter, other filters can be used to adjust the energy value of the frequency band that needs to be suppressed in the merged first filtered signal and the second filtered signal.

Different from FIG. 5 and FIG. 6 , energy band suppression can also be achieved inside the first transparent filter and the second transparent filter. For example, as shown in FIG. 7 , the suppression of the energy value of each frequency band can be directly achieved in the design of the first transparent filter and the second transparent filter. That is to say, the first transparent filter and the second transparent filter can not only achieve the transparency of the sound signal, but also suppress the energy value of each frequency band according to the corresponding adjustment amount. In this way, the first filtered signal output by the first transparent filter includes the low- and medium-frequency signals after energy band suppression, and the second filtered signal output by the second transparent filter includes the high-frequency signal after energy band suppression. The suppression of the energy values of the corresponding frequency bands is achieved simultaneously in the first transparent filter and the second transparent filter, without the need to set up additional hardware circuits, making the entire system structure more compact and reducing costs.

FIG8 shows a flowchart of a specific implementation of method 100. As shown in FIG8 , the environmental sound transparency stage may include hardware transparency and software transparency. In the hardware transparency part, a first transparency filter, i.e., a hardware transparency filter, is used to filter the sound signal collected by the external ear microphone 12, and output a first filtered signal mainly composed of medium and low frequency signals. Here, since the background noise in the medium and low frequency signals is very small, it can be ignored. In the software transparency part, the background noise signal is first estimated by software, and the background noise signal is removed from the sound signal collected by the external ear microphone 12. Then, a software transparency filter is used to filter the sound signal after the background noise is removed, and a second filtered signal mainly composed of high frequency signals is output. The background noise signal in the second filtered signal has been basically eliminated. The first filtered signal and the second filtered signal are filtered. By superimposing the wave signals, a cleaner compensation signal can be obtained.

As shown in FIG8 , in the adaptive processing stage of the ambient sound, the energy peak of each frequency band is estimated, and the frequency bands that need energy adjustment are determined. The adjustment amount g is determined based on the difference between the statistical value of the energy peak of each of these frequency bands and the corresponding energy threshold, that is, the statistical value minus the energy threshold. When the difference is greater than or equal to 0, that is, the statistical value-energy threshold ≥ 0, the adjustment amount is output. The energy peak of each frequency band can be counted in real time and the corresponding adjustment amount can be calculated until the difference gradually returns to 0. Among them, the energy of each frequency band in the compensation signal is adjusted according to the adjustment amount g, for example, the gain corresponding to each frequency band can be adjusted according to the corresponding adjustment amount. Finally, the microphone outputs the compensation signal after the adaptive processing stage.

The above-mentioned solution for adaptive processing of ambient sound can be implemented in combination with the aforementioned software and hardware combined transparent solution, or it can be implemented separately.

For example, as shown in FIG9 , the embodiment of the present application further provides a filtering method 200. The method 200 can be performed by a filtering device in the headset 10, for example, by a chip in the headset 10, such as a codec chip or a Bluetooth audio SOC. The headset 10 includes an external ear microphone 12 for collecting sound signals. As shown in FIG1 , the method 200 includes some or all of the following steps.

In step 210, energy values of multiple frequency bands of the sound signal are determined.

In step 220, the energy values of the multiple frequency bands of the sound signal are adjusted so that the energy values of the multiple frequency bands are less than or equal to energy thresholds corresponding to the multiple frequency bands respectively.

The multiple frequency bands may be, for example, frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the energy values of the multiple frequency bands of the sound signal are adjusted so that the energy values of the multiple frequency bands are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands, including: based on a transparent filter, the sound signal is processed to obtain a filtered signal; wherein, the energy values of the multiple frequency bands of the sound signal are adjusted so that the energy values of the multiple frequency bands are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands, including: the energy values of the multiple frequency bands of the filtered signal are adjusted so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, adjusting the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to energy thresholds corresponding to the multiple frequency bands respectively includes: based on the transparent filter or another filter other than the transparent filter, The energy values of the multiple frequency bands of the filtered signal are adjusted so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to energy thresholds corresponding to the multiple frequency bands respectively.

In one implementation, the energy values of the multiple frequency bands of the filtered signal are adjusted, including: determining an adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and an energy threshold corresponding to each frequency band, and adjusting the energy value of each frequency band of the sound signal based on the adjustment amount corresponding to each frequency band.

In one implementation, the adjustment amount corresponding to each frequency band is determined based on the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, including: determining the adjustment amount corresponding to each frequency band based on the difference between the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band.

It should be understood that the specific details of method 200 can be referred to the aforementioned description of the part regarding the adaptive processing of ambient sound in method 100, and for the sake of brevity, they will not be repeated here.

The present application also provides a filtering device 300, which is applied to an earphone 10 with a transparent mode, the earphone 10 includes an external ear microphone 12, and the sound signal collected by the external ear microphone 12 includes a signal of a first frequency band and a signal of a second frequency band, and at least part of the frequency band in the first frequency band is lower than at least part of the frequency band in the second frequency band. For example, the first frequency band is less than or equal to a preset first frequency threshold, and the second frequency band is greater than or equal to a preset second frequency threshold. The first frequency threshold and the second frequency threshold can be, for example, between 2KHz and 3KHz.

As shown in FIG. 10 , the device 300 includes a first transparent filter 310 , a processing module 320 , and a second transparent filter 330 .

Among them, the first transparent filter 310 is used to process the sound signal to obtain a first filtered signal of the sound signal, the first transparent filter is a hardware transparent filter, and the gain of the first transparent filter on the signal of the first frequency band in the sound signal is greater than the gain of the first transparent filter on the signal of the second frequency band in the sound signal; the processing module 320 is used to perform background noise reduction processing on the sound signal; the second transparent filter 330 is used to process the sound signal after background noise reduction to obtain a second filtered signal of the sound signal, the second transparent filter is a software transparent filter, and the gain of the second transparent filter on the signal of the second frequency band in the sound signal after background noise reduction is greater than the gain of the second transparent filter on the signal of the first frequency band in the sound signal after background noise reduction.

The software transparent filter may be, for example, a FIR filter or an IIR filter, and the hardware transparent filter includes, for example, a FIR filter or an IIR filter.

In one implementation, the processing module 320 is specifically used to: obtain a pre-estimated background noise signal; A sound signal after the background noise is reduced is determined according to the background noise signal and the sound signal.

In one implementation, the earphone 10 also includes a speaker 11 and an in-ear microphone 13, and the processing module 320 is also used to: determine the secondary transfer function based on the audio signal output by the speaker 11 and the in-ear signal collected by the in-ear microphone 13; determine the first transparent filter 310 based on the extra-ear signal collected by the extra-ear microphone 12 when the earphone 10 is worn and the signal received in the ear when the earphone 10 is not worn.

In one implementation, the processing module 320 is also used to: determine the energy values of the multiple frequency bands of the sound signal; and adjust the energy values of the multiple frequency bands of the compensation signal according to the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the compensation signal are less than or equal to the energy thresholds corresponding to the multiple frequency bands respectively.

In one implementation, the processing module 320 is specifically used to: determine the adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band; and adjust the energy value of each frequency band of the compensation signal based on the adjustment amount corresponding to each frequency band.

In one implementation, the processing module 320 is specifically configured to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the first transparent filter 310 is also used to adjust the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the first filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands; the first transparent filter 310 is also used to adjust the energy values of the multiple frequency bands of the sound signal after background noise reduction so that the energy values of the multiple frequency bands of the second filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the first transparent filter 310 is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal according to the adjustment amount corresponding to each frequency band; the second transparent filter 330 is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal after background noise reduction according to the adjustment amount corresponding to each frequency band.

In one implementation, the first transparent filter 310 is specifically configured to: The adjustment amount corresponding to each frequency band is determined by the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band; the second transparent filter 330 is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band.

In one implementation, the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, where N is a positive integer.

It should be understood that the specific details of the filtering device 300 can refer to the above description of the filtering method 100, and for the sake of brevity, they are not repeated here.

The present application also provides a filtering device 400 , which is applied to an earphone 10 with a transparent mode, and the earphone 10 includes an external ear microphone 12 for collecting sound signals. As shown in FIG. 11 , the filtering device 400 includes a processing module 410 and a filter module 420 .

The processing module 410 is used to obtain energy values of multiple frequency bands of the sound signal; the filter module 420 is used to adjust the energy values of multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands are less than or equal to the energy thresholds corresponding to the multiple frequency bands.

The multiple frequency bands may be, for example, frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, where N is a positive integer.

In one implementation, the filter module 420 includes a transparent filter, which is used to: process the sound signal to obtain a filtered signal; wherein the transparent filter is also used to: adjust the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands; or, the filter module also includes another filter other than the transparent filter, which is used to: receive the filtered signal and adjust the energy values of the multiple frequency bands of the filtered signal so that the energy values of the multiple frequency bands of the filtered signal are less than or equal to the energy thresholds corresponding to each of the multiple frequency bands.

In one implementation, the transparent filter or the other filter is specifically used to: determine the adjustment amount corresponding to each frequency band based on the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the compensation signal based on the adjustment amount corresponding to each frequency band.

In one implementation, the transparent filter or the other filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band.

It should be understood that the specific details of the filtering device 400 can refer to the description of the filtering method 200 above. For the sake of brevity, I will not go into details here.

The present application also provides a chip, including a processor and a memory, including a processor and a memory, the memory stores computer instructions, and the processor calls the computer instructions to enable the device to implement the apparatus according to the first aspect or any implementation of the first aspect. The chip can be, for example, a codec chip in the headset 10 or a Bluetooth audio SOC.

The present application also provides an earphone 10 having a transparent mode, and the earphone 10 includes: a speaker 11; an external microphone 12; an internal microphone 13; and a filtering device in any of the above embodiments, or a chip in any of the above embodiments.

It should be noted that, under the premise of no conflict, the various embodiments described in this application and/or the technical features in each embodiment can be arbitrarily combined with each other, and the technical solution obtained after the combination should also fall within the protection scope of this application.

The systems, devices and methods disclosed in the embodiments of the present application may be implemented in other ways. For example, some features of the method embodiments described above may be ignored or not performed. The device embodiments described above are merely schematic, and the division of units is merely a logical function division. There may be other division methods in actual implementation, and multiple units or components may be combined or integrated into another system. In addition, the coupling between the units or the coupling between the components may be direct coupling or indirect coupling, and the coupling may include electrical, mechanical or other forms of connection.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described devices and equipment and the technical effects produced can refer to the corresponding processes and technical effects in the aforementioned method embodiments, and will not be repeated here.

It should be understood that the specific examples in the embodiments of the present application are only to help those skilled in the art to better understand the embodiments of the present application, rather than to limit the scope of the embodiments of the present application. Those skilled in the art can make various improvements and modifications based on the above embodiments, and these improvements or modifications all fall within the scope of protection of the present application.

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 (24)

1. A filtering method, characterized in that it is applied to headphones with a transparent mode, the headphones include an external ear microphone, the sound signal collected by the external ear microphone includes a signal of a first frequency band and a signal of a second frequency band, at least part of the frequency band in the first frequency band is lower than at least part of the frequency band in the second frequency band, and the method includes:

   Processing the sound signal based on a first transparent filter to obtain a first filtered signal of the sound signal, wherein the first transparent filter is a hardware transparent filter, and a gain of the first transparent filter for the signal in the first frequency band of the sound signal is greater than a gain of the first transparent filter for the signal in the second frequency band of the sound signal;

   Performing noise reduction processing on the sound signal, and processing the sound signal after the noise reduction based on a second transparent filter to obtain a second filtered signal of the sound signal, wherein the second transparent filter is a software transparent filter, and the gain of the second transparent filter for the signal in the second frequency band in the sound signal after the noise reduction is greater than the gain of the second transparent filter for the signal in the first frequency band in the sound signal after the noise reduction;

   A compensation signal of the earphone is determined according to the first filtered signal and the second filtered signal.
2. The filtering method according to claim 1, characterized in that the noise reduction processing of the sound signal comprises:

   Obtaining a pre-estimated background noise signal;

   The sound signal after the background noise is reduced is determined according to the background noise signal and the sound signal.
3. The filtering method according to claim 1 or 2, characterized in that the earphone further comprises a speaker and an in-ear microphone, and the method further comprises:

   Determine a secondary transfer function according to the audio signal output by the speaker and the in-ear signal collected by the in-ear microphone;

   The first transparent filter is determined according to an external ear signal collected by the external ear microphone when the earphone is worn and a signal received in the ear when the earphone is not worn.
4. The filtering method according to any one of claims 1 to 3, characterized in that the method further comprises:

   determining energy values of a plurality of frequency bands of the sound signal;

   According to the energy values of the multiple frequency bands of the sound signal, the energy values of the multiple frequency bands of the compensation signal are adjusted respectively so that the energy values of the multiple frequency bands of the compensation signal are less than Or equal to the energy threshold corresponding to each of the multiple frequency bands.
5. The filtering method according to claim 4, characterized in that the energy values of the multiple frequency bands of the sound signal are adjusted respectively according to the energy values of the multiple frequency bands of the sound signal, comprising:

   Determine an adjustment amount corresponding to each frequency band according to an energy value of each frequency band of the sound signal and an energy threshold corresponding to each frequency band;

   The energy value of each frequency band of the compensation signal is adjusted according to the adjustment amount corresponding to each frequency band.
6. The filtering method according to claim 5, characterized in that the step of determining the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band comprises:

   The adjustment amount corresponding to each frequency band is determined according to the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band.
7. The filtering method according to any one of claims 1 to 3, characterized in that the method further comprises:

   determining energy values of a plurality of frequency bands of the sound signal;

   The first transparent filter is further used to adjust the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the first filtered signal are less than or equal to the energy thresholds corresponding to the multiple frequency bands respectively;

   The second transparent filter is further used to adjust the energy values of the multiple frequency bands of the sound signal after background noise reduction so that the energy values of the multiple frequency bands of the second filtered signal are less than or equal to energy thresholds corresponding to the multiple frequency bands respectively.
8. The filtering method according to claim 7, characterized in that:

   The first transparent filter is specifically used to determine, according to the energy value of each frequency band of the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, an adjustment amount corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal according to the adjustment amount corresponding to each frequency band;

   The second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal after background noise reduction according to the adjustment amount corresponding to each frequency band.
9. The filtering method according to claim 8, characterized in that:

   The first transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band;

   The second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band.
10. The filtering method according to any one of claims 4 to 9 is characterized in that the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, and N is a positive integer.
11. The filtering method according to any one of claims 1 to 10 is characterized in that the first frequency band is less than or equal to a preset first frequency threshold, the second frequency band is greater than or equal to a preset second frequency threshold, and the first frequency threshold and the second frequency threshold are between 2 kHz and 3 kHz.
12. A filtering device, characterized in that it is applied to headphones with a transparent mode, the headphones include an external ear microphone, the sound signal collected by the external ear microphone includes a signal of a first frequency band and a signal of a second frequency band, at least part of the frequency band in the first frequency band is lower than at least part of the frequency band in the second frequency band, and the filtering device includes:

    a first transparent filter, used to process the sound signal to obtain a first filtered signal of the sound signal, wherein the first transparent filter is a hardware transparent filter, and a gain of the first transparent filter for the signal in the first frequency band of the sound signal is greater than a gain of the first transparent filter for the signal in the second frequency band of the sound signal;

    A processing module, used for performing noise reduction processing on the sound signal;

    a second transparent filter, used for processing the sound signal after background noise reduction to obtain a second filtered signal of the sound signal, wherein the second transparent filter is a software transparent filter, and a gain of the second transparent filter for the signal of the second frequency band in the sound signal after background noise reduction is greater than a gain of the second transparent filter for the signal of the first frequency band in the sound signal after background noise reduction;

    The processing module is further configured to determine a compensation signal of the earphone according to the first filtered signal and the second filtered signal.
13. The filtering device according to claim 12 is characterized in that the processing module has Body is used for:

    Obtaining a pre-estimated background noise signal;

    The sound signal after the background noise is reduced is determined according to the background noise signal and the sound signal.
14. The filtering device according to claim 12 or 13, characterized in that the earphone further comprises a speaker and an in-ear microphone, and the processing module is further used for:

    Determine a secondary transfer function according to the audio signal output by the speaker and the in-ear signal collected by the in-ear microphone;

    The first transparent filter is determined according to an external ear signal collected by the external ear microphone when the earphone is worn and a signal received in the ear when the earphone is not worn.
15. The filtering device according to any one of claims 12 to 14, characterized in that the processing module is further used for:

    determining energy values of a plurality of frequency bands of the sound signal;

    According to the energy values of the multiple frequency bands of the sound signal, the energy values of the multiple frequency bands of the compensation signal are adjusted respectively so that the energy values of the multiple frequency bands of the compensation signal are less than or equal to the energy thresholds corresponding to the multiple frequency bands respectively.
16. The filtering device according to claim 15, characterized in that the processing module is specifically used for:

    Determine an adjustment amount corresponding to each frequency band according to an energy value of each frequency band of the sound signal and an energy threshold corresponding to each frequency band;

    The energy value of each frequency band of the compensation signal is adjusted according to the adjustment amount corresponding to each frequency band.
17. The filtering device according to claim 16, characterized in that the processing module is specifically used for:

    The adjustment amount corresponding to each frequency band is determined according to the difference between the energy value of each frequency band and the energy threshold corresponding to each frequency band.
18. The filter device according to any one of claims 12 to 14, characterized in that

    The first transparent filter is further used to adjust the energy values of the multiple frequency bands of the sound signal so that the energy values of the multiple frequency bands of the first filtered signal are less than or equal to energy thresholds corresponding to the multiple frequency bands respectively;

    The first transparent filter is further used to adjust the energy values of the multiple frequency bands of the sound signal after the background noise is reduced so that the energy values of the multiple frequency bands of the second filtered signal are less than Or equal to the energy threshold corresponding to each of the multiple frequency bands.
19. The filtering device according to claim 18, characterized in that

    The first transparent filter is specifically used to determine, according to the energy value of each frequency band of the multiple frequency bands of the sound signal and the energy threshold corresponding to each frequency band, an adjustment amount corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal according to the adjustment amount corresponding to each frequency band;

    The second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the energy value of each frequency band in the multiple frequency bands of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band, and adjust the energy value of each frequency band of the sound signal after background noise reduction according to the adjustment amount corresponding to each frequency band.
20. The filter device according to claim 19, characterized in that

    The first transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal and the energy threshold corresponding to each frequency band;

    The second transparent filter is specifically used to determine the adjustment amount corresponding to each frequency band according to the difference between the energy value of each frequency band of the sound signal after background noise reduction and the energy threshold corresponding to each frequency band.
21. The filtering device according to any one of claims 15 to 20 is characterized in that the multiple frequency bands are frequency bands whose energy values exceed corresponding energy thresholds among the N frequency bands of the sound signal, and N is a positive integer.
22. The filtering device according to any one of claims 12 to 21 is characterized in that the first frequency band is less than or equal to a preset first frequency threshold, the second frequency band is greater than or equal to a preset second frequency threshold, and the first frequency threshold and the second frequency threshold are between 2 kHz and 3 kHz.
23. A chip, characterized in that it comprises a processor and a memory, wherein the memory stores computer instructions, and the processor calls the computer instructions to enable the chip to implement the filtering method according to any one of claims 1 to 11.
24. A headset, characterized in that it has a transparent mode, comprising:

    speaker;

    External ear microphone;

    In-ear microphone; and,

    The filtering device according to any one of claims 12 to 22, or the chip according to claim 23.