WO2022198538A1

WO2022198538A1 - Active noise reduction audio device, and method for active noise reduction

Info

Publication number: WO2022198538A1
Application number: PCT/CN2021/082870
Authority: WO
Inventors: 张立斌
Original assignee: 华为技术有限公司
Priority date: 2021-03-25
Filing date: 2021-03-25
Publication date: 2022-09-29
Also published as: CN116982106A

Abstract

A method for active noise reduction, comprising: obtaining a first signal (302), the first signal representing first ambient sound acquired by a first microphone of an active noise reduction device during a first period of time; obtaining an evaluation signal corresponding to the first period of time (304); determining the length of a second period of time on the basis of the evaluation signal (306), the second period of time following the first period of time; and generating, on the basis of the first signal and the length of the second period of time, a second signal to be played by a first loudspeaker of the active noise reduction device (308), the second signal representing an inverse estimate of the first ambient sound during the second period of time. Further disclosed are a corresponding active noise reduction audio device and a computer readable storage medium. The active noise reduction audio device estimates inverse-phased sound at a subsequent moment on the basis of the current sound and plays the inverse-phased sound; and an estimated time length of the subsequent inverse-phased sound is adjusted by using a residual signal, so that the accuracy of inverse-phased signal estimation can be dynamically adjusted and a noise reduction effect can be improved.

Description

Active noise cancellation audio device and method for active noise cancellation

technical field

The present disclosure relates to the field of audio processing, and more particularly, to a method for noise reduction of ambient sound in an active noise reduction audio device and an active noise reduction audio device.

Background technique

As portable devices such as smartphones become more widely used, matching audio devices are also used more and more. When a user uses an audio device in a noisy environment, the user often wants to be able to filter or reduce the ambient sound so that the audio from the portable device can be clearly heard from the audio device, or simply want a quieter environment. environment of. At this time, even ambient sound such as music is equivalent to noise to the user.

Audio equipment generally has two noise filtering methods (noise reduction methods) to filter or reduce ambient sound. One form of noise reduction is passive noise reduction. Passive noise reduction is usually created by enclosing the ear to form a closed space, or using sound-isolating materials such as silicone earplugs to block outside noise. The noise reduction effect of passive noise reduction is limited, generally only high-frequency noise can be blocked, and the noise reduction effect of low-frequency noise is limited. Additionally, passive noise-cancelling audio devices can cause discomfort to the ears.

Another way to reduce noise is Active Noise Cancellation. Active noise reduction uses audio equipment to generate opposite-phase sound waves equal to the outside noise, neutralize the noise, and achieve the effect of noise reduction. For example, active noise-cancelling audio devices typically provide an ambient microphone at the outermost surface of the audio device, which is used to detect ambient sounds. Before the ambient sound reaches the inside of the ear, the audio equipment with active noise reduction needs to complete the detection of the ambient sound and the calculation and generation of the inverted signal. Ideally, when the inverted signal is accurately calculated and reaches the inside of the ear at the same time as the direct sound, better noise reduction can be achieved. However, limited by factors such as the calculation of the inverted signal, in some cases, the conventional active noise reduction audio equipment is still not ideal for the noise reduction of the ambient sound, and in some cases may also require high-cost circuit components .

SUMMARY OF THE INVENTION

In view of the above problems, the embodiments of the present disclosure aim to provide a technical solution for noise reduction of ambient sound.

According to a first aspect of the present disclosure, there is provided a method for active noise reduction. The method includes acquiring a first signal representing a first ambient sound collected by a first microphone of an active noise reduction device during a first period of time; and acquiring an evaluation signal corresponding to the first period of time. The method also includes determining, based on the evaluation signal, a length of time for a second time period, the second time period following the first time period. The method further includes generating a second signal to be played by the first speaker of the active noise reduction device based on the first signal and the time length of the second period. The second signal represents an inverse estimate of the first ambient sound during the second time period. The evaluation signal represents the effect of active noise reduction and/or the accuracy of the estimate. In one implementation, the evaluation signal can be obtained by a residual microphone described in detail later. Alternatively or additionally, in another implementation, the evaluation signal may be determined by the evaluation signal estimated by the processor during the initial period and the ambient sound collected in the first period. By estimating the reversed phase signal of the second period in advance and playing the reversed phase signal, the reversed phase sound can reach the human ear in synchronization with the direct sound at the subsequent time, shortening the delay of the reversed phase sound, and the reversed phase sound and the direct sound can be cancelled effectively. to improve the noise reduction effect. By using the residual signal to adjust the estimated time length of the subsequent anti-phase sound, the accuracy of the anti-phase signal estimation can be dynamically adjusted and the noise reduction effect can be improved. For example, when the residual signal indicates that the active noise reduction effect is better, the estimated time length can be extended. Conversely, when the residual signal indicates that the effect of active noise reduction is not ideal, the estimation time length can be shortened to improve the estimation accuracy.

In some implementations, generating the second signal to be played by the first speaker of the active noise cancellation device based on the first signal and the time length of the second time period includes generating the third signal based on the first signal and the time length of the second time period, The third signal represents an estimate of the first ambient sound during the second time period; and inverting the third signal to generate the second signal. The inverted signal can be obtained more accurately by estimating and then inverting to generate the inverted signal.

In some implementations, generating the second signal to be played by the first speaker of the active noise cancellation device based on the first signal and the time length of the second period includes inverting the first signal to generate a fourth signal, the fourth signal an inverted sound representing the first ambient sound during the first period; and generating the second signal based on the fourth signal and a time length of the second period.

In some implementations, obtaining the evaluation signal corresponding to the first time period includes obtaining, as the evaluation signal, a residual signal collected by a residual microphone of the active noise reduction audio device during the first time period. The residual microphone is different from the first microphone. By using the residual microphone to collect the residual sound as the evaluation signal, the effect of active noise reduction can be obtained more accurately and intuitively, so that the active noise reduction can be adjusted more effectively.

In some implementations, obtaining the evaluation signal corresponding to the first time period includes an inverting estimated signal corresponding to the first time period estimated based on the first signal and a time period preceding the first time period or based on the first signal and the first time period The estimated signal corresponding to the first period estimated for the period preceding the period determines the estimated signal. Alternatively, acquiring the evaluation signal corresponding to the first time period includes an estimated signal corresponding to the first time period estimated based on an inversion signal of the first signal and a time period preceding the first time period, or based on the inversion of the first signal The signal and the inverted estimated signal corresponding to the first period estimated for the period preceding the first period, determine the evaluation signal. Obtaining the residual signal in various ways can more comprehensively and accurately evaluate the effect of the current noise reduction.

In some implementations, generating the initial output signal to be played by the first speaker of the active noise cancelling audio device based on the initial input signal and the time length of the second time period includes determining the amount of the first ambient sound during the first time period based on the initial input signal a category; determining an estimated model corresponding to the category of the first ambient sound based on the category of the first ambient sound; and generating a second signal based on the estimated model, the time length of the second period, and the first signal. By classifying the ambient sound, the inverse phase estimation of the ambient sound can be more targeted. By using an estimation model that matches the ambient sound, the estimation accuracy can be improved and the noise reduction width and noise reduction depth can be increased accordingly.

In some implementations, generating the second signal based on the estimated model, the time length of the second time period, and the first signal includes: if the determined category of the ambient sound indicates that the ambient sound is speech, then using the speech model to perform analysis on the first signal estimation for generating the second signal; if the determined category of the ambient sound indicates that the ambient sound is noise, the neural network model is used to estimate the first signal for generating the second signal; if the determined ambient sound is of noise The class indicates that the ambient sound is music, using a reserve pool model to estimate the first signal for generating the second signal; and if the determined class of the ambient sound indicates that the ambient sound is a mixed sound, using a weighted model to evaluate the first signal. A signal is estimated for generating a second signal. By specifically classifying ambient sounds into speech, noise, music, and mixed sounds, and estimating them using corresponding speech models, neural network models, reserve pool models, and weighting models, respectively, the estimated signal can be obtained more accurately and The Active Noise Cancellation is improved accordingly.

In some implementations, if the speech is voiced, a reserve pool model is used to estimate the first signal for generating the second signal; and if the speech is unvoiced, a linear prediction model is used to estimate the first signal to for generating the second signal. By further subdividing unvoiced and voiced sounds, the estimation accuracy for speech can be further improved, and the effect of active noise reduction for speech is correspondingly improved.

In some implementations, the method further includes: adjusting an estimated model corresponding to the category of ambient sound based on the evaluation signal; and generating subsequent audio to be played by the first speaker based on the audio signal subsequently collected by the first microphone and the adjusted model Signal. For a single estimated model, adjusting the model includes replacing the single model and/or adding an estimated model. For multiple estimation models, adjusting the models includes replacing one or more models, adjusting model weights, and/or increasing or decreasing the estimated models. By using the residual signal to adjust the estimation model of the subsequent inverted sound, the accuracy of the estimation of the inverted signal can be dynamically adjusted and the noise reduction effect can be improved.

In some implementations, the method further includes: determining a filter corresponding to the ambient sound based on the category of the ambient sound; and generating a second signal based on the first signal and the filtering. By filtering the collected audio signal, the noise in the sound can be effectively filtered. Different filtering can be used for different sound classes. In addition, weighted filtering of various filters can also be used. Weighted filtering may have different filter weights for different sounds. Similarly, filtering can be adjusted based on feedback mechanisms such as the use of residual signals. Adjusting the filtering includes increasing/decreasing the filtering type, and/or adjusting the weight of each filter in the weighted filtering.

In some implementations, if the ambient sound is determined to be music, the first signal is filtered using a finite impulse response filter for generating the second signal. If the determined category of ambient sound indicates that the ambient sound is speech, the first signal is filtered using an infinite impulse response filter for generating the second signal. By using different filtering methods for different sounds, it is possible to filter out noise more effectively for voice and music, so as to obtain better active noise reduction effect.

In some implementations, determining the estimation model corresponding to the category of the first ambient sound based on the category of the first ambient sound includes determining a weighted model corresponding to the first ambient sound based on the category of the first ambient sound, the weighting model includes: a first The estimated model, the weight of the first estimated model, the second estimated model and the weight of the second estimated model. By using the weighted estimation model, the estimation of the ambient sound can be performed more effectively and in a more targeted manner, so as to correspondingly improve the noise reduction effect.

In some implementations, the method further includes adjusting the estimated model based on the residual signal. In one implementation, generating the second signal to be played by the first speaker based on the first signal and the time length of the second time period includes: generating the second signal to be played by the first speaker based on the first signal, the time length of the second time period, and the adjusted estimation model The second signal played by the first speaker. By using the residual signal to adjust the subsequent estimation model, an appropriate estimation model can be selected based on the noise reduction effect to improve the noise reduction effect.

In some implementations, determining the category of ambient sound during the first period based on the first signal includes at least one of a frame rate for the first energy range, a frame rate for the first frequency range, and a zero-crossing rate based on the first signal Determines the category of ambient sound. The first energy range includes the low energy range. The first frequency range includes the low frequency range. By identifying low-energy frame rates, low-frequency signal frame rates, and zero-crossing rates, sound signals can be effectively differentiated into speech, music, noise, silence, and mixed sounds. In this way, a suitable estimation model can be more accurately determined, thereby providing estimation accuracy, and correspondingly increasing the noise reduction width and noise reduction depth. In addition, speech can be further divided into unvoiced and voiced, thereby further improving the accuracy of speech estimation, and correspondingly increasing the noise reduction width and noise reduction depth.

In some implementations, generating the second signal to be played by the first speaker of the active noise cancellation device based on the first signal and the time length of the second period includes determining a filter corresponding to the first ambient sound based on the category of the first ambient sound ; generating a second signal based on the first signal, the time length of the second time period and the filtering. For different sound categories, it may contain different noises. By adopting corresponding filtering mechanisms for different types of sounds, noise can be filtered out more effectively.

In some implementations, the method further includes adjusting the filtering based on the evaluation signal. By using the evaluation signal to adjust subsequent filtering, an appropriate estimated filter can be selected based on the noise reduction effect to improve the noise reduction effect.

In some implementations, generating the initial output signal includes: using a weighted estimation model or a reserve pool model to estimate the first signal to generate an estimated signal, the estimated signal representing an estimate of ambient sound during the first time period; and evaluating the estimated signal Inversion is performed to generate the second signal. Alternatively, the first signal is inverted to generate an inverted signal, and a weighted estimation model or a pool model is used to estimate the inverted signal to generate the second signal. In addition, it is possible to dynamically adjust the coefficients of individual terms in the weighted model or to increase or decrease the number of models used by using the evaluation signal. By using the default model to estimate, the calculation amount of the estimation is reduced and the estimation speed is improved.

In some implementations, the method further includes acquiring a fifth signal, the fifth signal representing the second ambient sound collected by the second microphone in the active noise cancellation audio device during the first period; and based on the fifth signal and the second The length of time of the period generates a sixth signal to be played by the second speaker in the active noise cancellation audio device, the sixth signal representing an inverse estimate of the second ambient sound during the second period. In some implementations, different microphones have different collection effects for different frequency bands. For example, the first microphone can capture more details for low-frequency sounds, while the second microphone can capture more details for mid-high frequency sounds. Therefore, by using different microphones to adopt different collection schemes for the ambient sound, various details of the ambient sound can be collected more effectively. In addition, active noise cancellation can be ensured even if one microphone fails. In some implementations, different speakers have different performance effects in different frequency bands. For example, the first speaker is more prominent in the low frequency band, while the second speaker is more prominent in the mid-high frequency. Therefore, by using different speakers to play different inverted signals, different parts of the direct sound can be more accurately cancelled for better noise reduction depth and noise reduction width.

In some implementations, the method further includes first filtering the first signal to generate a filtered first filtered signal; generating a second signal based on the first filtered signal; Second filtering to generate a filtered second filtered signal; and generating a sixth signal based on the second filtered signal. By using multiple filters, multiple estimates, and playing multiple sounds, it is possible for different sound classes to use speakers tuned for that class for better noise reduction.

In some implementations, the method further includes: obtaining a first residual signal, the first residual signal representing the sum of the first sound and the direct sound, wherein the first sound is based on the initial output signal; obtaining a second residual a signal, the first residual signal represents the sum of the second sound and the direct sound, wherein the second sound is based on the sixth signal; the first filtering is adjusted based on the first residual signal; and the pair of The second filter is adjusted. By using a plurality of residual signals, the situation of inaccurate residual signal acquisition caused by the influence of the position or performance of the residual microphone can be avoided, so that a more stable noise reduction effect can be continuously obtained.

In some implementations, generating the second signal based on the first signal includes inverting the first signal to generate the first inverted signal; estimating the first inverted signal using the first estimation model to generate the first component signal ; use a second estimation model different from the first estimation model to estimate the first inverted signal to generate a second component signal different from the first component signal; use a weighting model to weight the first component signal and the second component signal , to generate the initial output signal. By using the default weighted model to estimate, the calculation amount of the estimation is reduced, the estimation speed is improved, and the estimation can be dynamically adjusted to improve the estimation accuracy.

According to a second aspect of the present disclosure, there is provided a computer-readable storage medium storing one or more programs. One or more programs are configured to be executed by one or more processors. The one or more programs comprise instructions for performing the method according to the first aspect. By reading and executing one or more programs in the computer-readable storage medium, the inverted signal of the second period can be estimated in advance. By playing the reversed-phase signal, the reversed-phase sound can reach the human ear synchronously with the direct sound at the subsequent time, shortening the time delay of the reversed-phase sound, and the reversed-phase sound and the direct sound can be canceled to effectively improve the noise reduction effect. By using the residual signal to adjust the estimated time length of the subsequent anti-phase sound, the accuracy of the anti-phase signal estimation can be dynamically adjusted and the noise reduction effect can be improved. For example, when the residual signal indicates that the active noise reduction effect is better, the estimated time length can be extended. Conversely, when the residual signal indicates that the effect of active noise reduction is not ideal, the estimation time length can be shortened to improve the estimation accuracy.

According to a third aspect of the present disclosure, there is provided a computer program product. A computer program product includes one or more programs. One or more programs are configured to be executed by one or more processors. The one or more programs comprise instructions for performing the method according to the first aspect. By executing the computer program product, the inverted signal of the second time period can be estimated in advance. By playing the reversed-phase signal, the reversed-phase sound can reach the human ear synchronously with the direct sound at the subsequent time, shortening the delay of the reversed-phase sound, and the reversed-phase sound and the direct sound can be canceled to effectively improve the noise reduction effect. By using the residual signal to adjust the estimated time length of the subsequent anti-phase sound, the accuracy of the anti-phase signal estimation can be dynamically adjusted and the noise reduction effect can be improved. For example, when the residual signal indicates that the active noise reduction effect is better, the estimated time length can be extended. Conversely, when the residual signal indicates that the effect of active noise reduction is not ideal, the estimation time length can be shortened to improve the estimation accuracy.

According to a fourth aspect of the present disclosure, an active noise reduction audio device is provided. The active noise reduction audio device includes an acquisition module for acquiring a first signal and an evaluation signal corresponding to a first time period, where the first signal represents the first ambient sound collected by the first microphone of the active noise reduction device during the first time period and an inverse estimation signal generation module for determining, based on the evaluation signal, a time length of a second time period, the second time period following the first time period; and A second signal played by the first speaker of the , the second signal representing an inverted estimate of the first ambient sound during the second time period. The ANC audio device may estimate the inverted signal for the second period in advance. By playing the reversed-phase signal, the reversed-phase sound can reach the human ear synchronously with the direct sound at the subsequent time, shortening the time delay of the reversed-phase sound, and the reversed-phase sound and the direct sound can be canceled to effectively improve the noise reduction effect. By using the residual signal to adjust the estimated time length of the subsequent anti-phase sound, the accuracy of the anti-phase signal estimation can be dynamically adjusted and the noise reduction effect can be improved. For example, when the residual signal indicates that the active noise reduction effect is better, the estimated time length can be extended. Conversely, when the residual signal indicates that the effect of active noise reduction is not ideal, the estimation time length can be shortened to improve the estimation accuracy.

In some implementations, the inverse estimation signal generation module is further configured to generate a third signal based on the first signal. The third signal represents an estimate of ambient sound during the second time period; and inverting the third signal to generate the second signal. The inverted signal can be obtained more accurately by estimating and then inverting to generate the inverted signal.

In some implementations, the acquisition module is further configured to acquire a first signal corresponding to the first microphone, where the first signal represents ambient sound collected during the first period of time. The inverse estimated signal generation module is further configured to determine the time length of the second period based on the first signal and the estimated inverse estimated signal of the previous period or the estimated signal based on the first signal and the estimated period of the previous period, and the second period is in the first period after. Alternatively, the inversion estimation signal generation module is further configured to determine the second phase based on the inversion signal of the first signal and the inversion estimation signal estimated in the previous period or based on the inversion signal of the first signal and the estimated signal estimated in the previous period. The time length of the period, the second period follows the first period. Obtaining the residual signal in various ways can more comprehensively and accurately evaluate the effect of the current noise reduction.

In some implementations, the obtaining module is further configured to obtain a residual signal corresponding to a residual microphone of the active noise reduction audio device, where the residual microphone is different from the first microphone. Obtaining the residual signal in various ways can more comprehensively and accurately evaluate the effect of the current noise reduction.

In some implementations, the inverse estimation signal generation module is further configured to determine a category of the first ambient sound during the first time period based on the first signal; determine an estimate corresponding to the category of the first ambient sound based on the category of the first ambient sound a model; and generating a second signal based on the estimated model, the first signal, and a time length of the second time period. By classifying the ambient sound, the inverse phase estimation of the ambient sound can be more targeted. By using an estimation model that matches the ambient sound, the estimation accuracy can be improved and the noise reduction width and noise reduction depth can be increased accordingly.

In some implementations, the inverse estimation signal generation module is further configured to determine a weighting model corresponding to the ambient sound based on the category of the ambient sound, where the weighting model includes: a first estimation model, a weight of the second estimation model, the first estimation model and The weights of the second estimation model. By using the weighted estimation model, the estimation of the ambient sound can be performed more effectively and in a more targeted manner, so as to correspondingly improve the noise reduction effect.

In some implementations, the inverse estimation signal generation module is also used to adjust the estimation model based on the estimation signal. By using the evaluation signal to adjust the subsequent estimation model, an appropriate estimation model can be selected based on the noise reduction effect to improve the noise reduction effect.

According to a fifth aspect of the present disclosure, an active noise reduction audio device is provided. The active noise cancellation audio device includes one or more processors; a memory storing one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including Instructions for the method of the first aspect. By using the residual signal to adjust the subsequent estimation model, an appropriate estimation model can be selected based on the noise reduction effect to improve the noise reduction effect. By using the residual signal to adjust the estimated time length of the subsequent anti-phase sound, the accuracy of the anti-phase signal estimation can be dynamically adjusted and the noise reduction effect can be improved. For example, when the residual signal indicates that the active noise reduction effect is better, the estimated time length can be extended. Conversely, when the residual signal indicates that the effect of active noise reduction is not ideal, the estimation time length can be shortened to improve the estimation accuracy.

According to a sixth aspect of the present disclosure, an active noise reduction audio device is provided. The active noise cancellation audio device includes a first microphone, one or more processors, and a first speaker. The first microphone is configured to collect the first ambient sound during the first period of time and generate the first signal. The one or more processors are configured to obtain an evaluation signal corresponding to the first time period; determine a time length of the second time period based on the evaluation signal, the second time period following the first time period; and the first signal and the time length of the second time period A second signal is generated, the second signal representing an inverted estimate of the first ambient sound during the second time period. The first speaker is configured to be configured to play the second signal during the second period of time. By using the residual signal to adjust the subsequent estimation model, an appropriate estimation model can be selected based on the noise reduction effect to improve the noise reduction effect. By using the residual signal to adjust the estimated time length of the subsequent anti-phase sound, the accuracy of the anti-phase signal estimation can be dynamically adjusted and the noise reduction effect can be improved. For example, when the residual signal indicates that the active noise reduction effect is better, the estimated time length can be extended. Conversely, when the residual signal indicates that the effect of active noise reduction is not ideal, the estimation time length can be shortened to improve the estimation accuracy.

In some implementations, the active noise reduction audio device further includes a residual microphone. The residual microphone is configured to acquire residual sound to generate a residual signal as an evaluation signal. By using the residual signal to adjust subsequent sound estimates, the estimates can be dynamically adjusted in a feedback manner for better dynamic active noise reduction.

In some implementations, the active noise reduction audio device further includes: a second microphone and a second speaker. The second microphone is configured to collect the second ambient sound during the first period of time and to generate a fifth signal that is different from the first signal. The second speaker is configured to play a sixth signal, wherein the sixth signal is generated by the one or more processors based on the fifth signal. The sixth signal represents a second inverted estimate of the second ambient sound during the second period and is different from the second signal.

It should be understood that the matters described in this Summary are not intended to limit key or critical features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Description of drawings

The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent when taken in conjunction with the accompanying drawings and with reference to the following detailed description. In the drawings, the same or similar reference numbers refer to the same or similar elements, wherein:

1 shows a schematic diagram of an active noise reduction audio device that may implement embodiments of the present disclosure;

2 shows a schematic block diagram of an active noise reduction audio device according to an embodiment of the present disclosure;

3 is a schematic flowchart of a method for filtering or reducing ambient sound according to an embodiment of the present disclosure;

4 is a schematic diagram of a method for classifying ambient sounds and performing inverse estimation according to an embodiment of the present disclosure;

5 is a schematic diagram of a process of active noise reduction according to an embodiment of the present disclosure;

6 is a schematic diagram of a process of active noise reduction according to another embodiment of the present disclosure;

7 shows a schematic diagram of another active noise cancellation audio device that may implement embodiments of the present disclosure;

8 is a schematic diagram of a process of active noise reduction according to yet another embodiment of the present disclosure;

9 is a schematic block diagram of a computer-readable storage medium according to one embodiment of the present disclosure; and

FIG. 10 is a schematic block diagram of an apparatus for denoising ambient sound according to an embodiment of the present disclosure.

Detailed ways

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for the purpose of A more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes, and are not intended to limit the protection scope of the present disclosure.

In the description of embodiments of the present disclosure, the term "comprising" and the like should be understood as open-ended inclusion, ie, "including but not limited to". The term "based on" should be understood as "based at least in part on". The terms "one embodiment" or "the embodiment" should be understood to mean "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. The term "period" refers to a continuous or discrete period of time, the smallest unit of which includes a single point in time. In other words, the term "period" may include at least one point in time. Other explicit and implicit definitions may also be included below. The term "representation" herein indicates that there is a direct or indirect relationship between the two, and that one can characterize a certain attribute and/or a change in attribute of the other. As used herein, "microphone" refers to a device for collecting sound and converting it into a corresponding electrical signal. "Speaker" refers to a device used to convert electrical signals into sound based on audio data. The term "ambient sound" herein refers to the sound present in the external environment in which the ANC audio device is located and captured by the ANC audio device, which may be a combination of one or more sounds, such as speech, music, noise, etc.

As mentioned above, in some situations, ambient sound may be noise to the user, and the user wishes to filter or reduce the ambient sound by using an active noise cancellation audio device. In this document, the filtering or reduction of ambient sound is collectively referred to as noise reduction of ambient sound. The existing active noise reduction and active noise reduction audio equipment is limited by factors such as the calculation of the inverted signal, and cannot filter or reduce the ambient sound well. For example, the time required for the real-time calculation of the inverted signal is greater than the propagation time of the ambient sound signal from the external sampling microphone of the ANC audio device to the inside of the ear, which can cause a mismatch between the inverted signal and the direct signal, and may even cause an increase in noise.

In response to the above problems and other potential problems, embodiments of the present disclosure provide a method for noise reduction of ambient sound, an active noise reduction audio device, and a computer-readable storage medium. In one embodiment of the present disclosure, ambient sound is collected using an ambient sound microphone of an active noise-cancelling audio device, an inverted signal representing a possible ambient sound for a subsequent period is predicted or estimated based on a signal representing the ambient sound, and The anti-phase signal is played using the built-in speaker of the active noise reduction audio device, so that the played anti-phase sound and the direct ambient sound cancel each other out at a subsequent moment, thereby realizing noise reduction of the ambient sound. More specifically, by using the scheme according to the embodiment of the present disclosure, a better noise reduction width can be obtained, that is, the frequency at which the effective noise reduction effect can be increased, compared to the conventional scheme of generating a real-time inverted signal to cancel the direct sound. width. At the same time, the solution according to the embodiment of the present disclosure can obtain a better noise reduction depth (for example, it can be expressed in dB), thereby improving the degree of effectively suppressed/cancelled noise. In addition, in one embodiment of the present disclosure, by acquiring the evaluation signal, the current effect of active noise reduction can be known, and the estimated length of the subsequent signal can be adjusted based on the evaluation signal representing the current active noise reduction effect, so as to obtain subsequent more accurate noise reduction. Accurate inverse signal estimation to improve the effect of subsequent active noise reduction.

FIG. 1 shows a schematic diagram of an active noise cancellation audio device 10 in which embodiments of the present disclosure may be implemented. In one embodiment, the active noise cancelling audio device 10 may be, for example, an ear-contact audio playback device such as a True Wireless Stereo (TWS) headset. The active noise cancellation audio device 10 may include a pair of earphones, and the two

earphones

11 and 12 are configured substantially identically to each other. Therefore only one earphone 11 is used for the schematic description. The headset 11 includes an external first microphone 13, a processor 17 located inside the headset 11, a first speaker located inside the headset 11 (relative to the external microphone 13 exposed to the environment) in an in-ear portion or in contact with the ear 15 and a residual microphone 14. The first microphone 13 is configured to detect or collect the sound of the external environment.

In some embodiments, the residual microphone 14 may not be present. In this case, dynamic adjustment of the estimates may not be required. Although a possible configuration of the active noise cancellation audio device 10 is shown in FIG. 1, this is for illustration only and does not limit the scope of the present disclosure. For example, in some embodiments, the two

earphones

11 and 12 may have only one processor 17, and the wireless signal is transmitted by wireless transmission such as Bluetooth signal transmission to realize the two

earphones

11 and 12 to a single processor 17 shares. In another embodiment, the two

earphones

11 and 12 may also share a single first microphone 13 .

In one embodiment, the external first microphone 13 of the active noise reduction audio device 10 collects ambient sound, and performs acousto-electrical conversion to generate a continuous electrical signal and transmit it to the processor 17 . The processor 17 predicts or estimates the ambient sound at the subsequent time based on the received signal, and generates and transmits to the first speaker 15 an inverted signal representing the ambient sound at the subsequent time. As used herein, "inverted signal" refers to a signal that is manipulated after inverting the audio signal, eg by directly inverting the sign of the audio sample points or further processing. In contrast, a signal that is not inverted may be referred to as a "non-inverting signal". The out-of-phase signal played by the speakers is used to cancel each other to a certain extent with the direct sound (normal-phase sound) directly inside the active noise cancelling audio device 10 to reduce the sound perceived inside the ear. The first speaker 15 plays the reversed-phase sound based on the received reversed-phase signal, so as to cancel the direct ambient sound from the environment directly into the active noise reduction audio device 10 at a subsequent time, so as to achieve the effect of noise reduction. Although the schematic configuration of the active noise cancellation audio device 10 is shown in FIG. 1 with TWS headphones, it is to be understood that the scope of the present disclosure is not so limited. For example, another possible configuration of an active noise cancelling audio device, ie over-ear headphones, is shown with headphones in Figure 7 below. Alternatively, the active noise reduction audio device 10 may also be, for example, an active noise reduction audio device that transmits audio through bone conduction.

FIG. 2 shows a schematic block diagram of an active noise reduction audio device 100 according to one embodiment of the present disclosure. It should be understood that the ANC audio device 100 shown in FIG. 2 is only exemplary, for example, used to illustrate one possible implementation of the ANC audio device 10 of FIG. 1 , and should not constitute a description of the present disclosure. any limitations on the functionality and scope of the implementation. In one embodiment, the ANC audio device 100 may include a processor 110, a wireless communication module 160, an antenna 1, an audio module 170, a speaker module 170A, a microphone module 170C, a key 190 shown in solid boxes and solid lines , an internal memory 121 , a universal serial bus (USB) interface 130 , a charging management module 140 , and a power management module 141 .

The processor 110 may be, for example, the processor 17 of FIG. 1 , and may include one or more processing units, for example, the processor 110 may include an application processor (application processor, AP), a modem processor, a graphics processor ( graphics processing unit (GPU), image signal processor (ISP), controller, video codec, digital signal processor (DSP), baseband processor and/or neural network processor ( neural-network processing unit, NPU), etc. In some embodiments, the different processing units may be separate devices. In other embodiments, different processing units may also be integrated in one or more processors. The controller can generate an operation control signal according to the instruction operation code and timing signal, and complete the control of fetching and executing instructions.

A memory may also be provided in the processor 110 for storing instructions and data. Internal memory 121 may be used to store computer executable program code, which includes instructions. The internal memory 121 may include a storage program area and a storage data area. The processor 110 executes various functional applications and data processing of the active noise reduction audio device 100 by executing instructions stored in the internal memory 121 and/or instructions stored in a memory provided in the processor.

The active noise reduction audio device 100 may implement audio functions through an audio module 170, a speaker module 170A, a microphone module 170C, an application processor, and the like. Such as music playback, recording, etc. The audio module 170 is used for converting digital audio information into analog audio signal output, and also for converting analog audio input into digital audio signal. Audio module 170 may also be used to encode and decode audio signals. In some embodiments, the audio module 170 may be provided in the processor 110 , or some functional modules of the audio module 170 may be provided in the processor 110 .

The speaker module 170A, also called "speaker", is used to convert audio electrical signals into sound signals. The active noise cancellation audio device 100 can listen to music through the speaker module 170A, or listen to hands-free calls. The microphone module 170C is also called "microphone" or "microphone". When making a call or sending a voice message, the user can make a sound by approaching the microphone module 170C through a human mouth, and input the sound signal to the microphone module 170C.

In other embodiments, the ANC audio device 100 may further include an antenna 2 and a mobile communication module 150 . In addition to the above components, the ANC audio device 100 may further include an external memory interface 120, a battery 142, an earphone 170B, a headphone interface 170D, a sensor module 180, a motor 191, an indicator 192, a camera shown in dashed and dashed boxes 193, a display screen 194, and one or more of a subscriber identification module (subscriber identification module, SIM) card interface 195. The sensor module 180 may include a pressure sensor 180A, a gyro sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, and an ambient light sensor One or more of 180L, bone conduction sensor 180M. Sensor module 180 may also include other types of sensors not listed.

It can be understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the active noise reduction audio device 100 . In other embodiments of the present application, the active noise reduction audio device 100 may include more or less components than shown, or combine some components, or separate some components, or arrange different components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

FIG. 3 is a schematic flowchart of a method 300 for noise reduction of ambient sound according to one embodiment of the present disclosure. In one embodiment, the method 300 may be performed by the active noise cancellation audio device 10 or the active noise cancellation audio device 100 . Where the active noise cancellation audio device 10 performs the method 300 , the processor 17 of the active noise cancellation audio device 10 may execute computer program code or instructions that implement the method 300 . Where the active noise cancellation audio device 10 performs the method 300 , the processor 110 of the active noise cancellation audio device 100 may execute computer program code or instructions that implement the method 300 . As an example only, the method 300 for noise reduction of ambient sound is described below by taking the active noise reduction audio device 10 as an example, and it should be understood that this is only exemplary and not limiting.

The ANC audio device 100 acquires an initial input signal corresponding to the first microphone 13 in the ANC audio device 10, the initial input signal representing the ambient sound collected by the ANC audio device 10 during the initial period. The initial input signal is acquired by, for example, the first microphone 13 and transmitted to the active noise reduction audio device 100 . It can be understood that the initial input signal may also be processed by an intermediate audio signal processing circuit (such as filtering, gain amplification, etc.) and then transmitted to the active noise reduction audio device 100 for subsequent processing. The scope of the embodiments of the present disclosure is not limited in this regard.

Ambient sounds are continuously generated, but from the perspective of the ANC audio device 10, the ANC audio device 10 usually processes the received signal with a fixed or variable data length, the fixed or variable data The length corresponds to the length of time of the duration of the sampled ambient sound. For example, the initial input signal received by the active noise cancellation audio device 10 includes a microphone signal for one or more sampling time points. Thus, in one embodiment, the active noise cancellation audio device 10 acquires and processes the microphone signal one time period at a time. For example, in one processing cycle of the ANC audio device 10, the ANC audio device 10 acquires and processes the microphone signal of the first period. In a subsequent processor cycle of the ANC audio device 10, the ANC audio device 10 acquires and processes the microphone signal for a first period after the initial period, and so on. The length of the initial period may or may not be the same as the length of the first period. For example, the first time period may include a first plurality of time points, and the initial time period may include a second, different number of time points.

The active noise cancellation audio device 10 generates an initial output signal to be played by the first speaker 15 in the active noise cancellation audio device 10 based on the initial input signal. The initial output signal represents an inverse estimate of the ambient sound during a first period following the initial period. In an embodiment of the present disclosure, the active noise reduction audio device 10 predicts or estimates ambient sound signals for successive periods (eg, a first period) based on the sound sample signals of a previous period (eg, an initial period), and then estimates or estimates ambient sound signals for the estimated ambient sound The signal is inverted to generate an initial output signal representative of sounds that the ambient sound may have produced during the first period of time. Alternatively, the initial input signal may also be first inverted to generate an inverted signal, and prediction or estimation based on the inverted signal may be performed to generate the initial output signal.

The first loudspeaker 15 can generate an inverted initial inverted sound inside the user's ear by playing the initial output signal, and at this time, the ambient sound actually generated in the first period of time directly reaches the inside of the ear. The initial anti-phase sound and the direct sound are added (because the anti-phase sound and the direct sound are in opposite phases, they actually cancel each other out to reduce the net volume), so as to obtain the effect of noise reduction. In this embodiment, by predicting or estimating the sound of the subsequent period, it is equivalent to prolonging the time left for sound processing, so that the processing circuit has enough time to generate the inverted signal and the effect of noise reduction is improved. On the other hand, since the processing circuit has sufficient time to generate the inverted signal, high-speed processing circuit is also not required, thereby reducing the circuit design complexity and manufacturing cost of the active noise reduction audio device 10 .

In one exemplary embodiment, the estimation based on the initial input signal may be performed using a reserve pool model or a weighting model, which will be described in detail below, to generate the initial output signal. For example, the ANC audio device 10 may directly invert the initial input signal to generate the seventh signal and estimate the seventh signal using a first estimation model, such as a speech model, to generate the first component signal and use a different A second estimation model, such as a neural network model, of the first estimation model estimates the seventh signal to generate a second component signal that is different from the first component signal. The ANC audio device 10 then weights the first component signal and the second component signal to generate an initial output signal using a weighting model, eg multiplying the first component signal by a first weighting factor to obtain a first product, the second component The signal is multiplied by a second weight factor to obtain a second product, and the first product and the second product are added to obtain an initial output signal. It can be understood that in the weighting model, the weighting coefficients for the first component signal and the second component signal can be dynamically adjusted. For example, the adjustment is performed by using the residual signal described later.

Other models may also be used to estimate the initial input signal, such as other neural network models, linear prediction models, etc., as long as the model can predict subsequent audio signals based on the current audio signal. It will be appreciated that in some embodiments, the model used may be dynamically replaced during the noise reduction of the ANC audio device 10, eg, the predictive model used by the ANC audio device 10 may in some cases change from A first model, such as a reserve pool model, switches to a second model, such as a linear prediction model. In other embodiments, the parameters or configuration of the model used can be adjusted dynamically. For example, the weighting coefficients of the various terms of the weighting model may be dynamically adjusted based on the frequency content or class of the audio signal. Alternatively, a default or fixed model can also be used to generate the initial output signal.

Since the environment in which the user wearing the active noise cancelling audio device 10 may be located may be diverse, and the user may travel through different environments, in some embodiments, the user may be in a category according to the environment in which the user is located or based on ambient sound. category to generate the initial output signal in a targeted manner, so as to obtain a better noise reduction effect. For example, the active noise reduction audio device 10 may automatically determine the category of the ambient sound based on the initial input signal collected by the first microphone 13 . Alternatively, the user may set the category of the environment through a portable electronic device in communication with the active noise cancelling audio device 10, or send a set command via the active noise cancelling audio device 10 to set the category of the environment by voice. In other embodiments, the user may set the level of noise reduction effect of the active noise cancellation audio device 10 . For example, the noise reduction effect of the active noise reduction audio device 10 is set to a low noise reduction level to avoid omitting the warning sound in the ambient sound.

At 302, the processor 17 acquires the first signal. The first signal represents the first ambient sound collected by the first microphone 13 of the active noise reduction device during the first period of time. The processing of the processor 17 here is similar to the processing of the initial input signal, and details are not repeated here. The corresponding descriptions regarding the initial input signal may apply here.

At 304, the processor 17 obtains an evaluation signal corresponding to the first time period. In the present disclosure, the evaluation signal represents the effect of active noise reduction and/or the accuracy of the estimate. In one embodiment, the evaluation signal can be obtained by a residual microphone described in detail later. Alternatively or additionally, in another embodiment, the evaluation signal may be determined by the evaluation signal estimated by the processor 17 in the initial period and the ambient sound collected in the first period. For example, the processor 17 estimates an initial estimation signal in the initial period, and the initial estimation signal represents a possible situation of the estimated ambient sound in the first period. The processor 17 receives the first signal collected by the first microphone 13 during the first period. The processor 17 then subtracts the initial estimated signal from the first signal, and the estimated signal can be determined. Alternatively, the processor 17 estimates an initial inverted signal during the initial period, and the initial inverted estimated signal represents the estimated inverted signal of a possible situation of the ambient sound in the first period, that is, the inverted estimated signal of the first signal . The processor 17 receives the first signal collected by the first microphone 13 during the first period. The processor 17 then adds the initial inverted estimate signal to the first signal, and the estimate signal can be determined. It can be understood that there may also be other processing methods to obtain the residual signal, such as subtracting and inverting the inverted signal of the first signal and the initial inverted estimated signal, or adding the inverted signal of the first signal to the initial estimated signal , which is not limited in the present disclosure.

At 306, the processor 17 determines, based on the evaluation signal, a length of time for a second time period, the second time period following the first time period. Since the processor 17 obtains the evaluation signal, the previous evaluation effect can be determined. Therefore, the processor 17 can adjust subsequent estimates accordingly based on the estimated effect, such as adjusting the time length of successive time periods, to obtain a more accurate estimate, thereby improving the effect of active noise reduction. For example, when the residual signal indicates that the estimation effect is good, the time length of the estimation period can be maintained or increased. When the residual signal signal indicates that the estimation effect is not ideal, the time length of the subsequent estimation period can be reduced. In addition, the residual signal can also be used to adjust, and/or add or subtract the estimation model, to adjust the weight coefficients of each item in the weighted model, and to adjust the filtering, as described later.

At 308, the processor 17 generates a second signal to be played by the first speaker 15 based on the first signal and the time length of the second period. The second signal represents an inverse estimate of the first ambient sound during the second time period. The first speaker 15 of the active noise cancellation device 10 may play the second signal. The processing of the processor 17 here is similar to the processing of the initial output signal, which is not repeated here. The corresponding descriptions regarding the initial output signal may apply here.

FIG. 4 is a schematic diagram of a method 400 for classifying ambient sounds and performing inverse estimation according to one embodiment of the present disclosure. In general, the ambient sound of the environment where the user is located can be classified into five categories: silence, speech, noise, music, and hybrid, where the hybrid includes, for example, a mixture of speech, music and/or noise. The active noise cancellation audio device 10 may determine a category of ambient sound based on the first signal, and generate a third signal based on the determined category and the first signal, and invert the third signal to generate a second signal. Alternatively, the first signal may be inverted first to generate the fourth signal, and the classification and estimation may be performed based on the inverted fourth signal to generate the second signal.

Although the ambient sounds are classified into the above five categories, this is for illustration only, and does not limit the scope of the present disclosure. For example, ambient sound can also be divided into three categories: music, noise, and silence. In this case, speech is classified as noise. Alternatively, in some cases, the classification may also be based on the frequency of the ambient sound. By classifying different types of ambient sound signals, the ambient sound can be noise-reduced more effectively to obtain a better noise-reduction effect.

In one embodiment, the active noise reduction audio device 10 may determine the category of the ambient sound based on at least one of a frame rate of the first energy range, a frame rate of the first frequency range, and a zero-crossing rate of the first signal. Specifically, at 402, the active noise cancellation audio device 10 acquires a first signal. After receiving the first signal, the active noise reduction audio device 10 performs characteristic analysis on the first signal to determine the classification of the first signal. For example, the first signal may be analyzed based on its energy frame rate and zero-crossing rate.

For the silent class, the silent class is the audio signal segment that the human ear cannot perceive, which may sometimes contain a small amount of noise. Therefore, in order to detect as accurately as possible, a short-term energy threshold and a short-term zero-crossing rate threshold can be used to judge the first signal. For example, at 404, the active noise reduction audio device 10 analyzes the first signal to determine whether its energy frame rate and zero-crossing rate are both above respective given thresholds. This approach is similar to endpoint detection. When both the short-term energy and the short-term zero-crossing rate of the first signal are less than the corresponding given thresholds, the ANC audio device 10 may determine that the first signal is a mute signal at 406 . This indicates that the environment is in a quiet state, and accordingly, at 408, the active noise cancellation audio device 10 need not generate the third signal and accordingly the second signal also need not be generated. This can save power consumption and prolong the use time of the active noise cancelling audio device 10 .

Returning to 404, when the ANC audio device 10 determines that at least one of the energy frame rate and the zero-crossing rate is not lower than the threshold, the ANC audio device 10 further analyzes at 410 whether the frame rate of the first energy range is higher than the threshold . In this embodiment, the first energy range is, for example, a low-energy range, and the first frequency range is, for example, a low-frequency range. It can be understood that the specific ranges of the low energy range and the low frequency range can be set accordingly according to the performance design of the active noise reduction audio device 10 . Alternatively, ambient sounds may also be classified with different energy ranges or frequency ranges, as long as the ambient sounds have corresponding characteristics in a particular energy and/or frequency range.

Speech signals and noise signals generally have higher low energy frame rates than music and mixed sounds. If the frame rate of the first energy range is higher than a given threshold, the active noise reduction audio device 10 may determine that the first signal representing the ambient sound is a speech signal or a noise signal. For speech signals and noise signals, although both have higher frame rates in the low energy range, they have different characteristics in terms of the frequency of the sound. For example, speech signals usually have more low-frequency frame rates, while noise signals are more dispersed in frequency due to their randomness, and have lower frame rates at low frequencies. Accordingly, at 412, the active noise reduction audio device 10 determines whether the frame rate of the first frequency range of the first signal is above a threshold. If above the threshold, the ANC audio device 10 determines at 413 that the first signal is a speech signal. The active noise reduction audio device 10 may then generate a third signal using the speech model for the speech signal. For example, at 415, the active noise reduction audio device 10 may generate a third signal based on the first signal using a speech model such as a reserve pool model or a linear prediction (LPC) model. It will be appreciated that other speech models than the reserve pool model or the linear prediction model may also be used.

Specifically, the alternating appearance of unvoiced and voiced sounds in a speech signal is a unique property, while the unvoiced part has a higher zero-crossing rate, and the voiced part has a lower zero-crossing rate, so the zero-crossing rate in the speech signal will change alternately . For the unvoiced signal, the reserve pool model has good estimation results, so it can be used to estimate the third signal for the unvoiced signal. The Reservoir model is also called the Echo state network. The calculation of the reserve pool model simplifies the training process of the network, solves the problems that the traditional recurrent neural network structure is difficult to determine and the training algorithm is too complicated, and also overcomes the memory fading problem of the recurrent network. As for the voiced signal, the inventors found that the LPC model has a better estimation effect, so it can be used to estimate the third signal for the voiced signal. In an example embodiment, the active noise reduction audio device 10 may use the reserve pool model and the LPC model alternately to estimate the speech signal. Alternatively, the active noise reduction audio device 10 may also use only the reserve pool model or the LPC model to estimate the speech signal. The accuracy of speech estimation can be further improved by using targeted models for unvoiced and voiced sounds.

Returning to 412 , if the array of the first frequency range is not above the threshold, the active noise cancellation audio device 10 determines at 414 that the first signal is a noise signal. In turn, the ANC audio device 10 may process the noise signal using a suitable model to generate a third signal. In one embodiment, a neural network model may be used to estimate the first signal to generate the third signal. Although estimated above using the reserve pool model, the LPC model and the neural network model for the speech signal and the noise signal to generate the third signal, this is for illustration only and does not limit the scope of the present disclosure. Other suitable models, such as the weighting model described below, may also be used to generate the third signal.

Returning to 410, if the first energy range is not above the given threshold, the active noise cancellation audio device 10 may determine that the ambient sound is music or a mixed sound. The characteristics of music are harmonious and pleasing to the ear, and the timbre or frequency range of the music is wide, and it can be music played by various musical instruments. Therefore, the audio components contained in the music are complex, and the energy value will be too large. In addition, music does not have the sudden change of unvoiced and voiced sounds of speech signals, and the change of energy value is not as severe as that of speech signals, so the low-energy frame rate value of music is lower. Similarly, mixed sound also has a lower low energy frame rate. Therefore, low energy frame rate is a useful feature for judging music and mixed sounds.

After this, the ANC audio device 10 further differentiates between music and mixed sounds. The mixed appearance of speech and music in life can often render an atmosphere. This mixed signal usually takes the music part as the background, and the speech part occupies the main part (that is, the speech energy) occupies the main part. The alternating appearance of unvoiced and voiced sounds in speech signals is a unique property, and the unvoiced part has a higher zero-crossing rate, while the voiced part has a lower zero-crossing rate, so the zero-crossing rate in the speech signal will change alternately. There is no alternation of clear and voiced sounds in the music signal, so the change of the zero-crossing rate is relatively stable, and an effective parameter to measure the change of the zero-crossing rate is the zero-crossing rate variance. Therefore, the music signal and the speech-music mixed signal can be distinguished by the variance of the zero-crossing rate. In one embodiment, the active noise reduction audio device 10 may differentiate between music and mixed sounds based on zero-crossing rate variance. For example, at 420, the active noise reduction audio device 10 analyzes the variance of the zero-crossing rate of the first signal to determine whether the variance of the zero-crossing rate of the first signal is above a given threshold.

If the zero-crossing rate variance is not above the given threshold, the active noise reduction audio device 10 determines at 422 that the first signal is a music signal. For the music signal, the inventors found that the reserve pool model also has a better estimation effect for the music signal. Accordingly, the active noise cancellation audio device 10 may use the reserve pool model to estimate the first signal at 424 to generate the third signal. Although a reserve pool model is used here for estimation based on the music signal, this is for illustration only and does not limit the scope of the present disclosure. In some embodiments, a weighted model may be used to estimate the third signal based on the music signal.

If the zero-crossing rate variance is above a given threshold, the ANC audio device 10 determines at 421 that the first signal is a mixed signal. For the mixed signal, the ANC audio device 10 may estimate the first signal at 423 using the weighted model to generate the second signal. In one embodiment, the active noise reduction audio device 10 may estimate the first signal using at least two models among the reserve pool model, the LPC model and the neural network model, and assign weight coefficients to the estimated results of each model. For example, if there are more voiced signals and less noise signals, a higher weight factor, such as 0.75, can be assigned to the result estimated by the LPC model for the voiced signal, and the result estimated by the neural network model for the noise signal can be assigned a higher weight coefficient, such as 0.75. The result is assigned a lower weight factor, such as 0.25. The corresponding estimation results are then multiplied by the weight coefficients, and the two products are added to obtain the final second signal. In another embodiment, the reserve pool model, the LPC model and the neural network model can also be used for weighted estimation to obtain the final third signal. It can be understood that the weighting coefficient for each model in the weighting model can be dynamically adjusted. For example, based on the residual signal described later, the active noise reduction audio device 10 can dynamically adjust each coefficient to obtain a better noise reduction effect.

It can be seen that by subdividing the classification of ambient sounds and using the estimation model suitable for each category, noise reduction can be effectively performed even if the ambient sound has a wide sound frequency range, thereby significantly increasing the Noise reduction width. On the other hand, ambient noises with strong decibels usually have specific sound characteristics and belong to specific categories. Therefore, by applying an estimation model in a targeted manner, the noise decibels can also be effectively reduced to obtain a significant noise reduction depth.

Although the classification process and the corresponding estimation process are sequentially described above according to the flow shown in FIG. 4 , it should be understood that this is for illustration only and does not limit the scope of the present disclosure. Other classification and estimation procedures are possible. For example, after acquiring the first signal, the active noise reduction audio device 10 may analyze various features of the first signal, and directly determine the classification category of the first signal based on the analysis result, without the need for step-by-step analysis as shown in FIG. 4 . to get the classification class of the first signal. Furthermore, in the case of more or fewer classes, corresponding model estimation approaches may be added or removed to more accurately generate the third signal and then invert to generate the second signal.

FIG. 5 is a schematic diagram of a process 500 of active noise reduction according to one embodiment of the present disclosure. After the first microphone 13 of the active noise reduction audio device 10 collects the ambient sound, it converts it into a first signal 502 and outputs it to the processor 17 . The processor 17 may selectively filter 520 the first signal 502 to generate a filtered first filtered signal 504 . In this context, filtering refers to operations performed on a signal other than estimation and inversion operations. Filtering 520 may include, for example, existing or future signal processing such as adjusting gain, band filtering, signal noise reduction, and the like. Alternatively, filtering 520 may also be absent in some embodiments. The processor 17 may then estimate 530 the first filtered signal 504, eg, using the method 300 of FIG. 3 and/or the process 400 of FIG. the first speaker 15. Note that, in FIG. 5, the estimation 530 includes the step of inversion, so the inversion operation is not repeated here. Although filtering 520 and estimating 530 are described above using processor 17 as an example, filtering 520 and estimating 530 may also be performed by other components. For example, filtering 520 may be performed by a separate filter. Furthermore, although it is shown in FIG. 5 that the filtering 520 is performed first and then the estimation 530 is performed, it is also possible to perform the estimation 530 first and then perform the filtering 520 .

The first speaker 15 plays the received second signal to generate the first sound 512 in the ear. At the same time, the direct sound 514 of the ambient sound during the second period also goes directly inside the ear via the active noise cancellation audio device 10 . Since the direct sound 514 is out of phase with the first sound 512 , the two actually cancel each other out such that the net volume of the sound 508 perceived inside the ear is reduced compared to the direct sound 514 and the first sound 512 . In this way, the active noise reduction audio device 10 can realize active noise reduction. In an embodiment of the present disclosure, since the inverted sound used for cancellation is the "predicted" inverted sound for the second time period, there is no need to provide a corresponding instant when the direct sound during the first time period reaches the inside of the ear The inverted sound of the first period of time. In this way, it is equivalent to shifting the generation of the reversed-phase sound by a period of time backward, thereby reducing the strict requirements on the processing speed of the processor 17, and avoiding or alleviating the direct sound and the reversed-phase sound caused by the same period of time. The problem of unsatisfactory noise reduction effect caused by the mismatch. On the other hand, hardware design complexity and correspondingly reduced hardware cost can also be reduced, since there is no need to use ultra-high-speed computing circuits in the active noise cancellation audio device 10 .

In one embodiment, the active noise cancellation audio device 10 also has a built-in residual microphone 14 . The residual microphone 14 is configured to pick up residual sound 516 inside the ear. Residual sound 516 is actually the residual sound after noise reduction. The residual microphone 14 thus also generates a residual signal 508 representing the residual sound 516 based on the acquired residual sound 516 and feeds the residual signal 508 back to the processor 17 as an evaluation signal. Processor 17 may adjust at least one of filtering 520 and estimating 530 based on residual signal 508 . For example, in one embodiment, the filtering 520 is adaptive filtering, which can automatically adjust the filtering 520 based on the residual signal 508 . Additionally, the processor 17 may also adjust the estimate 530 based on the residual signal 508 . For example, the length of time of a second period following the first period during which the processor 17 estimates the inverted audio signal of the sound collected for the second period may be adjusted. If the residual signal 508 is small, this indicates a better estimate. Therefore, the processor 17 can accordingly increase the time length of the second period based on the residual signal 508 , for example, increase the time points included in the second period from 1 to 2, 3 or 4. If the residual signal 508 is larger, this indicates that the estimation is less effective. Therefore, the processor 17 can correspondingly reduce the time length of the second period based on the residual signal 508, for example, reducing the estimated time points included in the second period from 4 to 3, 2, 1, or even 0 ( i.e., not estimated).

On the other hand, if the residual signal 508 is large, the processor 17 may also alter the estimation model or adjust the model parameters. For example, changing from an LPC model to a reserve pool model, or adjusting the parameters of a weighted model. In addition, the processor 17 can also adjust the filtering 520, because the unsatisfactory noise reduction effect may also be caused by the filtering 520. It can be understood that the processor 17 can adjust at least one of the above-mentioned second period length, estimated model, model parameters and filtering 520 to obtain better noise reduction performance. In addition, in the case where the noise reduction performance is set to be non-optimal noise reduction performance in order not to miss important ambient sounds (such as subway station announcement voice), the processor 17 may also adjust the length of the second time period, the estimation model, At least one of model parameters and filtering 520. For example, avoid the use of reserve pool models and LPC models or reduce their weighting factors in weighted models.

Although the residual microphone 14 and adaptive filtering and estimation adjustment based on the residual signal 508 are shown in FIG. 5, this is for illustration only and does not limit the scope of the present disclosure. In some embodiments, there may be no residual microphone 14, the filtering 520 is fixed filtering and the estimate 530 of the second signal is not adjusted accordingly. Furthermore, although only the first microphone 13 , the first speaker 15 and the residual microphone 14 are shown in FIG. 5 , this is for illustration only, and the active noise cancellation audio device 10 may have more microphones and/or speakers.

In some embodiments, the first microphone 13 may collect ambient sound during the first period of time to generate the first signal. The processor 17 may determine whether the previous estimate is accurate based on the previously estimated initial inversion estimation signal for the first period and the collected first signal representing the real ambient sound of the first period, and adjust the subsequent estimation based on the determined result. Estimated length of time. For example, the length of time for the second period following the first period is adjusted. If the sum of the first signal and the initial inverted estimated signal is small, this indicates that the estimation result is good, and the time length of subsequent estimation can be maintained or increased. Conversely, if the sum of the first signal and the initial inverted estimated signal is larger, it indicates that the estimation result is poor, and the time length of subsequent estimation can be correspondingly reduced. Alternatively, the processor 17 may also determine whether the previous estimate was accurate based on the previously estimated estimate signal for the first period and the first signal. If the difference between the first signal and the estimated signal is small, this indicates that the estimation result is better, and the time length of subsequent estimation can be maintained or increased. Conversely, if the difference between the first signal and the estimated signal is large, it indicates that the estimation result is poor, and the time length of subsequent estimation can be correspondingly reduced. Furthermore, similar to the residual signal, in addition to adjusting the time length of the subsequent estimation, the above-described manner can also be used to adjust the estimation model and/or filtering.

FIG. 6 is a schematic diagram of a process 600 of active noise reduction according to another embodiment of the present disclosure. After the first microphone 13 of the active noise reduction audio device 10 collects the ambient sound, it converts it into a first signal 602 and outputs it to the processor 17 . The processor 17 may perform a selective first filtering 630 on the first signal 602 to generate a filtered first filtered signal 604 . The first filtering 630 may include, for example, processing such as adjusting gain, band filtering, noise reduction, and the like. The processor 17 may then perform a first estimate 630 on the first filtered signal 604 , eg, using the method 300 of FIG. 3 and/or the process 400 of FIG. 4 , to generate and output a second signal 606 to the active noise cancellation audio device 10 The built-in first speaker 15. Note that in FIG. 6 , the first estimation 630 includes an inversion step, so the inversion operation will not be repeated here. The first speaker 15 plays the received second signal 606 to produce the first sound 612 in the ear.

FIG. 6 differs from FIG. 5 in that FIG. 6 also has a second branch to perform second filtering 622 and second estimation 632 . Similarly, the first microphone 13 outputs the first signal 602 to the processor 17 . Processor 17 may perform selective second filtering 622 on first signal 602 to generate filtered second filtered signal 605 . The second filtering 632 may include, for example, adjusting gain, band filtering, noise reduction, etc., and the second filtering 632 may be the same as or different from the first filtering 630 . The processor 17 may then perform a second estimate 632 on the second filtered signal 605 , eg, using the method 300 of FIG. 3 and/or the process 400 of FIG. 4 , to generate and output a sixth signal 607 to the active noise cancellation audio device 10 The built-in second speaker 18. Note that in FIG. 6 , the second estimation 632 includes an inversion step, so the inversion operation will not be repeated here. The second speaker 18 plays the received sixth signal 607 to produce a second sound 613 in the ear.

At the same time, the direct sound 614 of the ambient sound during the second period also goes directly inside the ear via the active noise cancellation audio device 10 . Since the direct sound 614 is in opposite phase to the first sound 612 and the second sound 613, the direct sound 614 and the first sound 612 and the second sound 613 actually cancel each other to a certain extent, so that the sound 616 is perceived inside the ear. The net volume is reduced compared to the direct sound 614 , the first sound 612 and the second sound 613 . In this way, the active noise reduction audio device 10 can realize active noise reduction.

In one embodiment, the second filter 622 is different from the first filter 620 and the second estimate 632 is different from the first estimate 630 . For example, the first filtering 620 may be for speech signals and the first estimation 630 is also for speech signals, while the second filtering 622 is for music signals and the second estimation 632 is also for music signals. As another example, the first filtering 620 is for low frequency audio signals and the first estimation 630 is also for low frequency signals, while the second filtering 622 is for medium and high frequency signals and the second estimation 632 is also for medium and high frequency signals. In this case, the optimal settings can be set accordingly for each class of signals. In addition, it can be understood that the corresponding first speaker 15 and the second speaker 18 can also be selected for different types of sounds, so as to obtain better noise reduction depth and noise reduction width, because the speakers for a specific type are often better than general-purpose speakers. The wide-range speakers have better sound tuning.

FIG. 7 shows a schematic diagram of another active noise cancellation audio device 20 in which embodiments of the present disclosure may be implemented. The active noise cancellation audio device 20 may be, for example, a headset. The active noise cancellation audio device 20 may include a pair of ear cup portions, and the two ear cup portions are configured substantially identically to each other. Therefore, only one ear cup part is schematically described. The ear cup part includes an external first microphone 13, a second microphone 19 and a processor 17 inside the ear cup part. The earcup portion also includes a first residual microphone 14, a second residual microphone 16, a first speaker 15, and a first residual microphone 14, a second residual microphone 16, a first speaker 15, and a second residual microphone 14 located inside the earcup portion (relative to the first microphone 13 and the second microphone 19 exposed to the environment). Two speakers 18. Both the first microphone 13 and the second microphone 19 are configured to detect or collect sounds of the external environment, and the first microphone 13 and the second microphone 19 may operate simultaneously or alternately and may collect the same or different sounds. In one embodiment, the first microphone 13 may have an internal first filter to pick up only sounds of the first frequency, and the second microphone 19 may have an internal second filter to pick up only sounds of the second frequency. For example, the first frequency is a low frequency, and the second frequency is a mid-high frequency. By capturing sounds for different frequencies, more ambient sound detail can be obtained for better sound estimation, resulting in better noise reduction width and noise reduction depth.

In one embodiment, the external first microphone 13 and the second microphone 19 of the active noise reduction audio device 20 collect ambient sound, and perform acousto-electrical conversion to generate a continuous electrical signal and transmit it to the processor 17 . The processor 17 predicts or estimates the ambient sound at the subsequent time based on the received signal, and generates and transmits an inverted signal representing the ambient sound at the subsequent time to the built-in first speaker 15 and second speaker 18 . The first speaker 15 and the second speaker 18 play the reversed-phase sound based on the received reversed-phase signal to cancel the direct ambient sound from the environment directly into the active noise reduction audio device 20 at a subsequent moment, so as to achieve the effect of noise reduction.

8 is a schematic diagram of a process 800 of active noise reduction according to yet another embodiment of the present disclosure. In one embodiment, process 800 may be implemented in the active noise cancellation audio device 20 shown in FIG. 7 . After the first microphone 13 of the active noise reduction audio device 20 collects the ambient sound during the first period of time, it is converted into a first signal 802 and output to the processor 17 . The processor 17 may perform a selective first filtering 820 on the first signal 802 to generate a filtered first filtered signal 804 . Filtering 820 may include, for example, adjusting gain, band filtering, noise reduction, and the like. The processor 17 may then perform a first estimate 830 on the first filtered signal 804, eg, using the method 300 of FIG. 3 and/or the process 400 of FIG. 4, to generate a second signal 806 corresponding to the second time period and output it to The built-in first speaker 15 of the active noise cancellation audio device 20 . Note that, in FIG. 8 , the estimation 830 includes an inversion step, so the inversion operation is not repeated here. Although the first filtering 820 and the first estimation 830 are described above with the processor 17 as an example, the first filtering 820 and the first estimation 830 may also be performed by other components. For example, the first filtering 820 may be performed by a separate filter. Furthermore, although it is shown in FIG. 8 that the first filtering 820 is performed first and then the first estimation 830 is performed, it is also possible to perform the first estimation 830 first and then perform the first filtering 820 .

Similarly, after the second speaker 18 of the active noise cancellation audio device 20 has picked up the ambient sound during the first period, it is converted into a fifth signal and output to the processor 17 . The processor 17 may perform a selective second filtering 822 on the fifth signal to generate a filtered second filtered signal 805 . The second filtering 822 may include, for example, processing such as adjusting gain, band filtering, noise reduction, and the like. The processor 17 may then perform a second estimate 832 on the second filtered signal 805, eg, using the method 300 of FIG. 3 and/or the process 400 of FIG. 4, to generate a sixth signal 807 corresponding to the second time period and output it to The built-in second speaker 18 of the active noise cancellation audio device 20 . Note that in FIG. 8 , the second estimation 832 includes an inversion step, so the inversion operation will not be repeated here.

The first speaker 15 plays the received second signal 806 to produce the first sound 823 in the ear, and the second speaker 18 plays the received sixth signal 807 to produce the second sound 825 in the ear. The first sound 823 and the second sound 825 may be played simultaneously or alternately. The direct sound 824 of the ambient sound during the second period also goes directly inside the ear via the active noise cancellation audio device 20 . Since the direct sound 824 is out of phase with the first sound 823 and the second sound 825, the direct sound 824 and the first sound 823 and the second sound 825 actually cancel each other to a certain extent, so that the sound 826 and The net volume of 827 is reduced compared to the direct sound 824 , the first sound 823 and the second sound 825 . In this way, the active noise reduction audio device 20 can implement active noise reduction.

In one embodiment, the active noise reduction audio device 20 also has a built-in first residual microphone 14 and a second residual microphone 16 . The residual microphone 14 is configured to pick up the first residual sound 826 inside the ear. The first residual sound 826 is actually the residual sound after noise reduction. The first residual microphone 14 thus also generates a first residual signal 808 representative of the first residual sound 826 based on the acquired first residual sound 826 and feeds back the first residual signal 808 to the processor 17 . Processor 17 may adjust at least one of first filtering 820 and first estimation 830 based on first residual signal 808 . For example, in one embodiment, the first filtering 820 is adaptive filtering, which can automatically adjust the first filtering 820 based on the first residual signal 808 . Furthermore, the processor 17 may also adjust the first estimate 830 based on the first residual signal 808 . For example, the length of time of the second period following the first period during which the processor 17 estimates the inverted audio signal for the second period based on the sound signal collected during the first period may be adjusted. If the first residual signal 808 is small, this indicates a better estimation effect. Therefore, the processor 17 can correspondingly increase the time length of the second period based on the first residual signal 808 , for example, increase the time points included in the second period from 1 to 2, 3 or 4. If the first residual signal 808 is larger, this indicates that the estimation effect is less ideal. Therefore, the processor 17 can correspondingly reduce the time length of the second period based on the first residual signal 808 , for example, reduce the estimated time points included in the second period from 4 to 3, 2, 1, or even 0 (ie, not estimated).

On the other hand, if the first residual signal 808 is large, the processor 17 may also change the estimation model or adjust the model parameters. For example, changing from an LPC model to a reserve pool model, or adjusting the parameters of a weighted model. In addition, the processor 17 may also adjust the first filter 820, because the unsatisfactory noise reduction effect may also be caused by the first filter 820. It can be understood that the processor 17 can adjust at least one of the above-mentioned second period length, estimated model, model parameters and first filtering 820 to obtain better noise reduction performance. In addition, in the case where the noise reduction performance is set to be non-optimal noise reduction performance in order not to miss important ambient sounds (such as subway station announcement voice), the processor 17 may also adjust the length of the second time period, the estimation model, At least one of a model parameter and a first filter 820. For example, avoid the use of reserve pool models and LPC models or reduce their weight coefficients in weighted models.

Similarly, the processor 17 may adjust at least one of the length of the second period, the estimated model, the model parameters, and the second filtering 822 based on the second residual signal 810 to obtain better noise reduction performance. Although the first residual microphone 14, the second residual microphone 16, and the corresponding adaptive filtering and estimation adjustments based on the first residual signal 808 and the second residual signal 810 are shown in FIG. 8, this is only are illustrative and not limiting of the scope of the present disclosure. In some embodiments, the first residual microphone 14 and/or the second residual microphone 16 may not be present, eg, the first filtering 820 and/or the second filtering 822 are fixed filtering and the first estimate for the second signal 806 830 and/or the second estimate 832 for the sixth signal 807 is not adjusted accordingly. Additionally, it is understood that the active noise cancellation audio device 20 may have more microphones and/or speakers.

Although it is shown in FIG. 8 that the first sound 823 and the direct sound 824 and the second sound 825 and the direct sound 824 are canceled respectively, this is only a schematic example when the first sound 823 and the second sound 825 are played alternately. When the first sound 823 and the second sound 825 are played at the same time, the first sound 823, the second sound 825 and the direct sound 824 can be added together, and the same residual sound is generated for the first residual microphone 14 and the second residual sound Difference microphone 16 captures. In this case, the first residual signal 808 and the second residual signal 810 may be the same. Alternatively, the first residual signal 808 and the second residual signal 810 may also be different in this case, eg due to the performance and location of the first residual microphone 14 and the second residual microphone 16 . In this case, the processor 17 may average the first residual signal 808 and the second residual signal 810 for use in the first filtering 820 , the second filtering 822 , the first estimation 830 and the second estimation 832 . This avoids noise-canceling disturbances due to location or microphone performance, providing a more stable noise-canceling effect.

In an embodiment of the present disclosure, since the inverted sound used for cancellation is the "predicted" inverted sound for the second time period, there is no need to provide a corresponding instant when the direct sound during the first time period reaches the inside of the ear The inverted sound of the first period of time. In this way, it is equivalent to shifting the generation of the reversed-phase sound by a period of time backward, thereby reducing the strict requirements on the processing speed of the processor 17, and avoiding or alleviating the direct sound and the reversed-phase sound caused by the same period of time. The mismatch of noise reduction performance leads to the problem of unsatisfactory performance. On the other hand, hardware design complexity and correspondingly reduced hardware cost can also be reduced because ultra-high-speed computing devices are not required.

It will be appreciated that the first filter 820 and the second filter 822 may be the same or different, and the first estimate 830 and the second estimate 832 may be the same or different. In one embodiment, the second filter 822 is different from the first filter 820 , and the second estimate 832 is different from the first estimate 830 . For example, the first filtering 820 may be for speech signals and the first estimation 830 is also for speech signals, while the second filtering 822 is for music signals and the second estimation 832 is also for music signals. As another example, the first filtering 820 is for low frequency audio signals and the first estimation 830 is also for low frequency signals, while the second filtering 822 is for medium and high frequency signals and the second estimation 832 is also for medium and high frequency signals. Accordingly, the first sound 823 is a low frequency sound, and the second sound 825 is a medium and high frequency sound. In this case, the optimal settings can be set accordingly for each class of signals. In addition, it can be understood that the corresponding first speaker 15 and the second speaker 18 can also be selected for different types of sounds, so as to obtain better noise reduction depth and noise reduction width, because the speakers for a specific type are often better than the general ones. Wide-range speakers have better sound tuning.

FIG. 9 is a schematic block diagram of a computer-readable storage medium 900 according to one embodiment of the present disclosure. The computer-readable storage medium 900 is, for example, a cache in the processor 17, the internal memory 121 in FIG. 2, and the like. The computer readable storage medium 900 stores one or more programs 902 . . . 906 configured to be executed by one or more processors of an active noise cancellation audio device. One or more of the programs 902 . . . 906 may individually or collectively include instructions that may be executed by the processor 17 to implement the methods or processes described herein, such as the method 300 shown in FIG. The method 400 shown in FIG. 5 , the process 500 shown in FIG. 5 , the process 600 shown in FIG. 6 , and/or the process 800 shown in FIG. 8 . It will be appreciated that the computer-readable storage medium 900 may also include programs for implementing other methods and steps.

FIG. 10 is a schematic block diagram of an apparatus 1000 for denoising ambient sound according to an embodiment of the present disclosure. The apparatus 1000 may be applied to active noise reduction audio equipment. The apparatus 1000 includes: an acquisition module and an inversion estimation signal generation module. The acquisition module is used for acquiring the first signal and the evaluation signal corresponding to the first period. The first signal represents the first ambient sound collected by the first microphone of the active noise reduction device during the first period of time. an inverse estimation signal generation module for determining, based on the evaluation signal, a time length of a second time period, the second time period following the first time period; and generating a first signal to be used by the active noise reduction device based on the first signal and the time length of the second time period A second signal played by the speaker, the second signal representing an inverse estimate of the first ambient sound during the second time period. By predicting the sound of the future time based on the audio signal at the current time, better noise reduction width and depth of noise reduction can be obtained, and the circuit design can also be simplified and the demand for high-speed processing circuits can be reduced, thereby reducing the active noise reduction audio equipment. cost.

Although only two modules are shown in Figure 10, it is understood that this is illustrative only and not limiting of the scope of the present disclosure. Apparatus 1000 may also include corresponding modules for performing various steps in method 300 , method 400 , process 500 , process 600 and/or process 800 described above.

Although the subject matter has been described in language specific to structural features and/or logical acts of method, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

A method for active noise cancellation, comprising:

acquiring a first signal, the first signal representing the first ambient sound collected by the first microphone of the active noise reduction device during the first period of time;

obtaining an evaluation signal corresponding to the first time period;

determining a length of time for a second time period based on the evaluation signal, the second time period following the first time period; and

A second signal to be played by a first speaker of the active noise reduction device is generated based on the first signal and a time length of the second time period, the second signal representing the presence of a first ambient sound during the second time period Inverse estimation.
The method of claim 1, wherein generating a second signal to be played by a first speaker of the active noise reduction device based on the first signal and a time length of a second time period comprises:

generating a third signal based on the first signal and a time length of the second time period, the third signal representing an estimate of the first ambient sound during the second time period; and

The third signal is inverted to generate the second signal.
The method of claim 1, wherein generating a second signal to be played by a first speaker of the active noise reduction device based on the first signal and a time length of a second time period comprises:

inverting the first signal to generate a fourth signal representing an inverted sound of the first ambient sound during the first time period; and

The second signal is generated based on the fourth signal and the time length of the second period.
The method of any one of claims 1-3, wherein obtaining an evaluation signal corresponding to the first period of time comprises:

A residual signal collected by a residual microphone of the active noise reduction audio device during the first period of time is acquired as the evaluation signal, the residual microphone being different from the first microphone.
The method of any one of claims 1-3, wherein obtaining an evaluation signal corresponding to the first period of time comprises:

An inverted estimated signal corresponding to the first period estimated based on the first signal and a period preceding the first period or estimated based on the first signal and a period preceding the first period , an estimated signal corresponding to the first time period, and the estimated signal is determined.
5. The method of any of claims 1-5, wherein generating a second signal to be played by a first speaker of the active noise reduction device based on the first signal and a time length of a second time period comprises:

determining a category of a first ambient sound during the first time period based on the first signal;

determining an estimation model corresponding to the category of the first ambient sound based on the category of the first ambient sound; and

The second signal is generated based on the estimated model, the first signal, and a time length of a second time period.
The method of claim 6, wherein determining an estimation model corresponding to the category of the first ambient sound based on the category of the first ambient sound comprises:

A weighting model corresponding to the first environmental sound is determined based on the category of the first environmental sound, where the weighting model includes: a first estimation model, a weight of the first estimation model, a second estimation model and a weight of the second estimation model Weights.
The method according to claim 6 or 7, further comprising:

The estimation model is adjusted based on the estimation signal.
The method of any of claims 6-8, wherein generating a second signal to be played by a first speaker of the active noise reduction device based on the first signal and a time length of a second time period comprises:

determining a filter corresponding to the first ambient sound based on the category of the first ambient sound;

The second signal is generated based on the first signal, the time length of the second period, and the filtering.
The method according to any one of claims 1-9, further comprising:

obtaining a fifth signal representing a second ambient sound collected by a second microphone in the active noise reduction audio device during the first period of time; and

A sixth signal to be played by a second speaker in the ANC audio device is generated based on the fifth signal and a time length of the second time period, the sixth signal representing the noise during the second time period an inverse estimate of the second ambient sound.
A computer-readable storage medium storing one or more programs, the one or more programs being configured to be executed by one or more processors, the one or more programs comprising for executing the Instructions for any of the methods.
A computer program product comprising one or more programs configured to be executed by one or more processors, the one or more programs comprising means for executing claim 1 - Instructions for the method of any of -10.
An active noise-cancelling audio device comprising:

an acquisition module, configured to acquire a first signal and an evaluation signal corresponding to a first time period, the first signal representing the first ambient sound collected by the first microphone of the active noise reduction device during the first time period; and

Inverted estimation signal generation block for

determining a length of time for a second time period based on the evaluation signal, the second time period following the first time period; and

A second signal to be played by a first speaker of the active noise reduction device is generated based on the first signal and a time length of the second time period, the second signal representing the presence of a first ambient sound during the second time period Inverse estimation.
The active noise reduction audio device of claim 13, wherein the acquisition module is further configured to:

An inverted estimated signal corresponding to the first period estimated based on the first signal and a period preceding the first period or estimated based on the first signal and a period preceding the first period , an estimated signal corresponding to the first time period, and the estimated signal is determined.
The active noise cancellation audio device of claim 13, wherein

The obtaining module is further configured to obtain, as the evaluation signal, a residual signal collected by a residual microphone of the active noise reduction audio device during the first period of time, where the residual microphone is different from the first microphone.
The active noise reduction audio device according to any one of claims 13-15, wherein the inversion estimation signal generation module is further configured to:

determining a category of a first ambient sound during the first time period based on the first signal;

determining an estimation model corresponding to the category of the first ambient sound based on the category of the first ambient sound; and

The second signal is generated based on the estimated model, the first signal, and a time length of a second time period.
The active noise reduction audio device of claim 16, wherein the inverse estimation signal generation module is further configured to:

A weighted model corresponding to the first environmental sound is determined based on the category of the first environmental sound, where the weighted model includes: a first estimation model, a weight of the first estimation model, a second estimation model and a weight of the second estimation model Weights.
17. The active noise reduction audio device of claim 16 or 17, wherein the inverse estimation signal generation module is further adapted for the estimation signal to adjust the estimation model.
An active noise-cancelling audio device comprising:

one or more processors;

a memory storing one or more programs configured to be executed by the one or more processors, the one or more programs comprising means for performing any of claims 1-10 instructions for the method described.
An active noise-cancelling audio device comprising:

The first microphone, configured as:

collecting a first ambient sound during a first period of time and generating a first signal;

One or more processors, configured as:

obtaining an evaluation signal corresponding to the first time period;

determining a length of time for a second time period based on the evaluation signal, the second time period following the first time period; and

the first signal and the time length of the second time period generate a second signal representing an inverted estimate of the first ambient sound during the second time period; and

A first speaker configured to play the second signal during the second period of time.
The active noise cancellation audio device of claim 20, further comprising:

A residual microphone configured to collect residual sound to generate the residual signal as the evaluation signal.