WO2022027208A1 - 主动降噪方法、主动降噪装置以及主动降噪系统 - Google Patents

主动降噪方法、主动降噪装置以及主动降噪系统 Download PDF

Info

Publication number
WO2022027208A1
WO2022027208A1 PCT/CN2020/106681 CN2020106681W WO2022027208A1 WO 2022027208 A1 WO2022027208 A1 WO 2022027208A1 CN 2020106681 W CN2020106681 W CN 2020106681W WO 2022027208 A1 WO2022027208 A1 WO 2022027208A1
Authority
WO
WIPO (PCT)
Prior art keywords
audio signal
user
signal
target
noise reduction
Prior art date
Application number
PCT/CN2020/106681
Other languages
English (en)
French (fr)
Inventor
张立斌
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2020/106681 priority Critical patent/WO2022027208A1/zh
Priority to CN202080005894.7A priority patent/CN114391166A/zh
Publication of WO2022027208A1 publication Critical patent/WO2022027208A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering

Definitions

  • the present application relates to the field of active noise reduction, and in particular, to an active noise reduction method, an active noise reduction device, and an active noise reduction system.
  • ANC Active noise cancellation
  • any external audio signal is suppressed in the active noise reduction system; however, in general, users still want to effectively perceive the audio information they are concerned about; for example, the audio signal that the user is interested in or needs-related. Real-time audio signal.
  • the present application provides an active noise reduction method, an active noise reduction device, and an active noise reduction system, so that the noise reduction signal after the active noise reduction process still includes the target audio signal concerned by the user, which can satisfy the user's personality demand.
  • an active noise reduction method including:
  • the target audio signal is used to represent audio information that the user is concerned about
  • the ambient audio signal is used to represent the audio information of the environment where the user is located
  • the initial reference signal is used to indicate that active noise reduction processing is performed on the ambient audio signal
  • a target reference signal is obtained, and the target reference signal does not include the target an audio signal
  • the noise reduction signal is obtained according to the target reference signal and the ambient audio signal, and the noise reduction signal is used to cancel the target reference signal.
  • the target audio signal may be different for different users.
  • the target audio signal may refer to an audio signal that the user pays attention to; for example, the target audio signal may refer to an audio signal in the ambient audio signal that the user is interested in, or an audio signal related to the user's needs in the ambient audio signal.
  • the method of acquiring the user's target audio signal may include, but is not limited to: obtaining the audio information corresponding to the user's neuron state when the user's neuron state meets the preset conditions through the user's neuron state reflected by the acquired user's brain wave data, that is, the user's neuron state.
  • the audio information that the user pays attention to can be obtained by obtaining the user's behavior log; wherein, the user's behavior log can include the user's hobbies, the user's download history information, the user's browsing record information, and the like.
  • the ambient audio signal may refer to the audio information in the environment that the user perceives when the noise reduction device is used and the noise reduction function is not turned on;
  • the initial reference signal may refer to the audio collected by the device in the noise reduction device. Signal; for example, it may refer to the audio signal collected by the microphone of the noise reduction device. It should be understood that there may be differences between the audio information included in the ambient audio signal and the initial reference signal. Among them, any device or device that assists the noise reduction function can be considered as a part of the noise reduction device.
  • the target reference signal is obtained by acquiring the target audio signal that the user pays attention to; the target reference signal does not include the target audio signal of the user; according to the target reference signal and the ambient audio signal, the noise reduction signal is obtained, and the noise reduction signal is obtained.
  • the signal is used to cancel the target reference signal; so that the noise reduction signal after the active noise reduction process still includes the target audio information that the user pays attention to, so that the active noise reduction method can meet the user's individual needs.
  • the acquiring the user's target audio signal includes:
  • the target audio signal of the user is acquired according to the brain wave data of the user.
  • the target audio signal that the user pays attention to can be obtained according to the obtained brain wave data of the user on the environmental audio signal.
  • brain wave data may also be referred to as brain wave data, which refers to the data of the user's brain obtained by using electrophysiological indicators to record brain activity.
  • brain wave data refers to the data of the user's brain obtained by using electrophysiological indicators to record brain activity.
  • the post-synaptic potentials generated by a large number of neurons synchronously are summed to form brain waves, which record the changes in the electrical waves during brain activity and are the overall reflection of the electrophysiological activities of brain nerve cells on the surface of the cerebral cortex or scalp.
  • Brainwave data can be different for different users.
  • the brainwave data is used to reflect the neuron state of the user when the ambient audio signal is acquired, and the target audio signal refers to the The audio signal in which the neuron state of the user satisfies the preset condition in the ambient audio signal.
  • the preset condition may include that the fluctuation range of the neuron state of the user satisfies a preset threshold, or other preconfigured conditions; for example, the audio signal that causes the neuron state to fluctuate greatly may be the target audio signal.
  • the target audio signal refers to an audio signal preconfigured according to the user's behavior log.
  • the user's behavior log interests and hobbies the user's download history information, the user's browsing record information, and the like.
  • the user's target audio signal may be acquired according to preconfigured user information.
  • the target audio signal may refer to preconfigured text information or voice information that the user pays attention to.
  • the acquiring the user's target audio signal according to the user's brainwave data includes:
  • the vocal tract motion information is used to represent the motion information of the vocal tract occlusal part when the user speaks;
  • the target audio signal is obtained according to the channel motion information.
  • a large amount of vocal tract motion information when a person speaks can be associated with the user's brain wave data (for example, neural activity data);
  • the channel operation information can include the motion information of lips, tongue, larynx and mandible;
  • the audio information corresponding to the channel motion information can be obtained by obtaining a pre-trained recurrent neural network through the training data; wherein, the training data can include the sample channel motion information and the sample channel motion information.
  • Audio information the recurrent neural network is used to establish the correlation between the channel running information and the audio information.
  • the target audio signal is the target audio signal at the current moment, and further includes:
  • the target audio signal at the next moment at the current moment is predicted according to the target audio signal at the current moment.
  • the current moment and the next moment of the current moment may be continuous moments, or may also refer to discontinuous moments; for example, the next moment of the current moment may differ from the current moment by a periodic time interval, and the time interval
  • the time unit can include milliseconds or microseconds.
  • obtaining the user's target audio signal through brainwave data may not meet some real-time service requirements; therefore, a prediction model can also be used to obtain the user's target audio signal; the prediction model is used based on historical moments or The brainwave data of the user at the current moment and the target audio signal of the user predict the target audio signal of the user at the future moment.
  • the obtaining the target reference signal according to the target audio signal and the initial reference signal includes:
  • the initial reference signal is filtered according to the target audio signal to obtain the target reference signal.
  • performing filtering processing on the initial reference signal according to the target audio signal includes:
  • Filter processing is performed on the initial reference signal by using a filter according to the target audio signal.
  • the above-mentioned filters may include adaptive filters, or other filters.
  • the acquiring an ambient audio signal includes:
  • the ambient audio signal is received from a collaboration device, the collaboration device being configured to acquire the ambient audio signal from a sound source.
  • the user by using the cooperative device to forward the collected ambient audio signals, the user can perceive these signals before the ambient audio signals reach the user; thus, the target of the user's attention can be obtained more efficiently audio signal.
  • the collaboration device may send the ambient audio signal to the noise reduction device, and before playing the ambient audio signal on the noise reduction device, the energy level of the ambient audio signal may be scaled by the noise reduction device,
  • the playing sound does not affect the normal perception of the user; it is used to trigger the user to perceive the audio information of interest in advance, so as to form this audio perception before the arrival of the direct sound wave, and present the target audio information based on the brain wave data.
  • the collected ambient audio signal may also be scaled by the cooperation device and then forwarded to the noise reduction apparatus.
  • an active noise reduction device including:
  • the acquisition module is used to acquire the target audio signal, the environmental audio signal and the initial reference signal of the user, wherein the target audio signal is used to represent the audio information that the user is concerned about, and the environmental audio signal is used to represent the user's audio information.
  • the audio information of the environment, the initial reference signal is used to perform active noise reduction processing on the environmental audio signal;
  • the processing module is used to obtain a target reference signal according to the target audio signal and the initial reference signal, and the The target reference signal does not include the target audio signal;
  • the noise reduction signal is obtained according to the target reference signal and the ambient audio signal, and the noise reduction signal is used to cancel the target reference signal.
  • the obtaining module is specifically configured to:
  • the processing module is specifically used for:
  • the target audio signal of the user is acquired according to the brain wave data of the user.
  • the brainwave data is used to reflect the neuron state when the user acquires the ambient audio signal
  • the target audio signal refers to the ambient audio In the signal
  • the user's neuron state satisfies the audio signal of the preset condition.
  • the target audio signal refers to an audio signal preconfigured according to the user's behavior log.
  • the processing module is specifically configured to:
  • the vocal tract motion information is used to represent the motion information of the vocal tract occlusal part when the user speaks;
  • the target audio signal is obtained according to the channel motion information.
  • the target audio signal is the target audio signal at the current moment
  • the processing module is further configured to:
  • the target audio signal at the next moment at the current moment is predicted according to the target audio signal at the current moment.
  • the processing module is specifically configured to:
  • the initial reference signal is filtered according to the target audio signal to obtain the target reference signal.
  • the processing module is specifically configured to:
  • the initial reference signal is filtered by using an adaptive filter according to the target audio signal.
  • the obtaining module is specifically configured to:
  • the ambient audio signal is received from a cooperating device, the cooperating device is configured to acquire the ambient audio signal from a sound source.
  • an active noise reduction device comprising a memory for storing a program; a processor for executing the program stored in the memory, and when the program stored in the memory is executed, the processor is used for executing: acquiring a user's target audio signal, an environmental audio signal, and an initial reference signal, wherein the target audio signal is used to represent the audio information that the user is concerned about, and the environmental audio signal is used to represent the audio information of the environment where the user is located,
  • the initial reference signal is used to perform active noise reduction processing on the ambient audio signal; a target reference signal is obtained according to the target audio signal and the initial reference signal, and the target reference signal does not include the target audio signal ; the noise reduction signal is obtained according to the target reference signal and the ambient audio signal, and the noise reduction signal is used to cancel the target reference signal.
  • the above-mentioned active noise reduction apparatus includes a processor further configured to execute the first aspect and the active noise reduction method in any one of the implementation manners of the first aspect.
  • an active noise reduction earphone is provided, which is used to perform the active noise reduction method in the first aspect and any one of the implementation manners of the first aspect.
  • a vehicle headrest device for performing the active noise reduction method in the first aspect and any one of the implementation manners of the first aspect.
  • an automobile including an active noise reduction device implementing the second aspect and any one of the implementations of the second aspect.
  • an active noise reduction system for implementing the second aspect and the active noise reduction apparatus in any one of the implementation manners of the second aspect.
  • a computer program product comprising instructions, when the computer program product is run on a computer, the computer is made to execute the above-mentioned first aspect and the active noise reduction method in any one of the implementation manners of the first aspect .
  • a chip in a ninth aspect, includes a processor and a data interface, the processor reads an instruction stored in a memory through the data interface, and executes any one of the first aspect and the first aspect An active noise reduction method in an implementation.
  • the chip may further include a memory, in which instructions are stored, the processor is configured to execute the instructions stored in the memory, and when the instructions are executed, the The processor is configured to execute the first aspect and the active noise reduction method in any one of the implementation manners of the first aspect.
  • 1 is a schematic block diagram of an active noise reduction system
  • Figure 2 is a schematic diagram of the principle of the active noise reduction system
  • 3 is a schematic diagram of the superposition and cancellation of the noise reduction signal and the noise signal
  • FIG. 4 is a schematic diagram of the basic principle of an active noise reduction system
  • FIG. 5 is a schematic flowchart of an active noise reduction method provided by an embodiment of the present application.
  • FIG. 6 is an architectural diagram of an active noise reduction method provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram of obtaining an ambient audio signal through a collaborative device provided by an embodiment of the present application.
  • FIG. 8 is a schematic diagram of a method for acquiring a target audio signal based on user brainwave data
  • FIG. 9 is a schematic diagram of filtering an audio signal
  • FIG. 10 is a schematic block diagram of an active noise reduction device provided by an embodiment of the present application.
  • FIG. 11 is a schematic block diagram of an active noise reduction device provided by an embodiment of the present application.
  • FIG. 12 is a schematic diagram of a hardware structure of an active noise reduction device provided by an embodiment of the present application.
  • Active noise cancellation is a technology based on the principle of superposition of sound waves to achieve noise removal through mutual cancellation of sound waves; active noise cancellation systems can include feedforward and feedback.
  • composition and noise reduction principle of the active noise reduction system will be described in detail below with reference to Figure 1, Figure 2, Figure 3 and Figure 4.
  • the active noise reduction system 100 may generally include a controller 110 , a speaker 120 , an error mic 130 , and a ref mic 140 .
  • FIG. 2 the working principle and workflow of the active noise reduction system shown in FIG. 1 are described in detail.
  • Step 1 The error sensor 130 collects the error signal e(n), and transmits the error signal e(n) to the controller 110 .
  • the error signal e(n) is used to represent the sound field characteristics in the quiet zone shown in FIG. 2; for example, the sound field characteristics include sound pressure characteristics.
  • the sound field characteristics include sound pressure characteristics.
  • the error sensor 130 may include an acoustic sensor, an optical sensor or other sensors; the error sensor 130 may acquire audio information by collecting vibration.
  • the error sensor 130 may include a microphone.
  • Step 2 The reference sensor 140 collects the noise signal x(n), and transmits the noise signal x(n) to the controller 110 .
  • the noise signal x(n) may refer to the initial reference signal used in the present application to perform active noise reduction processing on the ambient audio signal, that is, the noise signal x that may be collected by the reference sensor in the embodiment of the present application.
  • the noise signal x(n) collected by the reference sensor 140 is an ambient noise signal. Ambient noise signals are often emitted by undesired noise sources, as shown in Figure 2.
  • Reference sensor 140 is typically an acoustic sensor. As shown in FIGS. 2 and 3 , the reference sensor 140 is a microphone.
  • Step 3 The controller 110 calculates an error cost function based on the error signal e(n), and predicts the noise reduction signal y(n) output by the speaker 120 based on the noise signal x(n) based on the principle of minimizing the error cost function.
  • the noise reduction signal y(n) is used to cancel the noise signal x(n).
  • the noise reduction signal y(n) is the inverted signal of the noise signal x(n), that is, the noise reduction signal y(n) and the noise signal x(n) are added to zero.
  • the noise reduction signal y(n) may also be referred to as an anti-noise signal.
  • the controller 110 may be an adaptive filter.
  • Step 4 the speaker 120 outputs the noise reduction signal y(n) according to the control of the controller 110; for example, the controller 110 calculates the error cost function of the superimposed signal e(n) after reaching the quiet zone; and based on the error cost function The minimum principle controls the output signal of the speaker 120 .
  • the speaker 120 controls the noise reduction signal y(n) output by the speaker according to the reverse signal generated by the controller 110 for canceling the noise. As shown in Figure 2, the noise signal x(n) and the noise reduction signal y(n) reach the dead zone through the primary path and the secondary path, respectively.
  • the error sensor 130 collects is the superimposed sound signal after the noise signal x(n) and the noise reduction signal y(n) pass through the primary path and the secondary path respectively and reach the quiet zone.
  • the sound signal is called is the error signal e(n).
  • the noise signal e(n) collected by the error sensor 130 can also be described as residual noise after noise reduction processing.
  • the goal of the controller 110 predicting the noise reduction signal y(n) output by the speaker 120 is to make the noise signal x(n) and the noise reduction signal y(n) pass through the primary path and the secondary path respectively and reach the quiet zone after the superimposed signal
  • the error cost function of e(n) is the smallest.
  • the speaker 120 may be referred to as the secondary sound source, as shown in FIG. 2 .
  • the superposition effects of the noise signal and the noise reduction signal at different positions are not necessarily the same.
  • the error sensor collects the error signal at point A, which can represent the superposition effect of the noise signal and the noise reduction signal at point A, but not necessarily the superposition effect of the noise signal and the noise reduction signal at other positions other than point A .
  • the concept of quiet zone is proposed, which represents the area or space where the error signal collected by the error sensor is located. That is, where the error sensor collects the signal, there is the quiet zone.
  • the dead zone represents the region where the error signal e(n) collected by the error sensor 130 is located.
  • the controller 110 to predict the noise reduction signal y(n) output by the speaker 120 is to make the noise reduction signal y(n) and the noise signal x(n) respectively reach the quiet zone and the superposed signal e( n) has the smallest error cost function. That is to say, the active noise reduction system aims to achieve the active noise reduction effect in the quiet zone.
  • FIG. 4 is a schematic diagram of the basic principle of an active noise reduction system.
  • the above-mentioned active noise reduction system may refer to a comprehensive active noise reduction system, and the comprehensive active noise reduction system is a combination of a feedforward type and a feedback type.
  • d may refer to a signal source signal, including the sum of useful signal S and noise interference n 0 ;
  • the essence of the active noise reduction system is to remove the signal containing x(n) or related to x(n) in d(n). If the content of x(n) is basically consistent with d(n), then The d(n) signal is cancelled. .
  • the ambient audio signal d(n) which can also be called the signal source signal
  • the noise signal x(n) in the collected environment will not be differentiated, that is, the noise signal x(n) is fully noise-reduced to the greatest extent, and any external sound signal is suppressed.
  • the current active noise reduction methods cannot meet the needs of users. Personalized requirements, that is, it is impossible to perform selective noise reduction processing according to the needs of users.
  • the present application proposes an active noise reduction method and an active noise reduction device, which obtain a target reference signal by acquiring a target audio signal that the user pays attention to; wherein, the target reference signal does not include the user's target audio signal; according to The target reference signal and the ambient audio signal are used to obtain a noise reduction signal, and the noise reduction signal is used to offset the target reference signal; so that the noise reduction signal after the active noise reduction process still includes the target audio signal that the user is concerned about, so that the active noise reduction method can Meet the personalized needs of users.
  • the active noise reduction method of the present application can be applied to a noise reduction device, wherein the noise reduction device refers to a device with noise reduction requirements; user headrest device.
  • the noise-cancelling playback device may include active noise-cancelling headphones, active noise-cancelling headband equipment, and the like;
  • the user headrest device in the vehicle scene may refer to a device for playing sound at the seat headrest in a car.
  • FIG. 5 is a schematic flowchart of an active noise reduction method provided by an embodiment of the present application.
  • the execution subject of the method 200 is any one of the above noise reduction apparatuses.
  • the method 200 includes steps S210 to S230, and the steps S210 to S230 are described in detail below respectively.
  • the target audio signal is used to represent the audio information that the user is concerned about
  • the environmental audio signal is used to represent the audio information of the environment where the user is located
  • the initial reference signal is used to perform active noise reduction processing on the environmental audio signal.
  • the target audio signal may refer to the audio signal that the user pays attention to; for example, the target audio signal may refer to the audio signal of the user's interest in the ambient audio signal, or the audio signal related to the user's needs in the ambient audio signal.
  • the target audio signal can be obtained by acquiring the user's brainwave data, and the target audio information that the user is concerned about can be determined according to the brainwave data reflecting the neuron state of the user; or, the target audio signal can also be pre-configured according to the user's behavior log Audio information; the user's behavior log may include the user's hobbies, the user's download history information, and the like.
  • the ambient audio signal may refer to the audio information in the environment that the user perceives when the noise reduction device is used and the noise reduction function is not turned on;
  • the initial reference signal may refer to the audio collected by the device in the noise reduction device. Signal; for example, it may refer to the audio signal collected by the microphone of the noise reduction device. It should be understood that there may be differences between the audio information included in the ambient audio signal and the initial reference signal. Among them, any device or device that assists the noise reduction function can be considered as a part of the noise reduction device.
  • acquiring the user's target audio signal includes: acquiring the user's brainwave data; and acquiring the user's target audio signal according to the user's brainwave data.
  • brain wave data may also be referred to as brain wave data, which refers to the data of the user's brain obtained by using electrophysiological indicators to record brain activity.
  • brain wave data refers to the data of the user's brain obtained by using electrophysiological indicators to record brain activity.
  • the post-synaptic potentials generated by a large number of neurons synchronously are summed to form brain waves, which record the changes in the electrical waves during brain activity and are the overall reflection of the electrophysiological activities of brain nerve cells on the surface of the cerebral cortex or scalp.
  • Brainwave data can be different for different users.
  • the brainwave data is used to reflect the neuron state of the environmental audio signal acquired by the user, and the target audio signal refers to the target audio signal in the environmental audio signal.
  • the user's neuron state satisfies the audio signal of the preset condition.
  • the preset condition may include that the fluctuation range of the neuron state of the user satisfies a preset threshold, or other preconfigured conditions; for example, the audio signal that causes the neuron state to fluctuate greatly may be the target audio signal.
  • the user's brainwave data can be converted into vocal tract motion information, and the vocal tract motion information is used to represent the motion information of the vocal tract occlusal part when the user speaks; the target audio signal is obtained according to the vocal tract motion information.
  • the target audio signal that the user pays attention to may be acquired according to the user's brainwave data.
  • the specific process please refer to the schematic diagram of the method for acquiring target audio information based on user brainwave data shown in the subsequent FIG. 8 .
  • the target audio signal refers to an audio signal preconfigured according to the behavior log of the user.
  • the user's behavior log includes the user's hobbies, the user's download history information, and the like.
  • the target audio signal is the target audio signal at the current moment, and further includes:
  • the target audio signal at the next moment at the current moment is predicted according to the target audio signal at the current moment.
  • obtaining the user's target audio signal through brainwave data may not meet some real-time service requirements; therefore, a prediction model can also be used to obtain the user's target audio information; the prediction model is used based on historical moments or The brain wave data of the user at the current moment and the target audio information of the user predict the target audio information of the user at the future moment.
  • the current moment and the next moment of the current moment may be continuous moments, or may also refer to discontinuous moments; for example, the next moment of the current moment may differ from the current moment by a periodic time interval, and the time interval
  • the time unit can include milliseconds or microseconds.
  • acquiring an ambient audio signal includes:
  • An ambient audio signal is received from the collaborative device, and the collaborative device is used to obtain the ambient audio signal from the sound source.
  • the user by using the cooperative device to forward the collected ambient audio signals, the user can perceive these signals before the ambient audio signals reach the user; thus, the target of the user's attention can be obtained more efficiently audio information.
  • the specific process please refer to the following figure 7 .
  • the target reference signal does not include the target audio signal.
  • the target reference signal is obtained according to the target audio signal and the initial reference signal, including:
  • the initial reference signal is filtered according to the target audio signal to obtain the target reference signal.
  • the initial reference signal is used to represent the noise signal that performs active noise reduction processing on the ambient audio signal, that is, the initial reference signal includes the audio signal that the user pays attention to and the audio signal that the user does not pay attention to; After filtering, the obtained target reference signal can filter out the audio signal that the user does not pay attention to.
  • performing filtering processing on the initial reference signal according to the target audio signal including:
  • Filter processing is performed on the initial reference signal by using a filter according to the target audio signal.
  • the filter may include an adaptive filter, or other filter; the specific process can be referred to as shown in the subsequent FIG. 9 .
  • the noise reduction signal is used to cancel the target reference signal.
  • the noise reduction signal can generate a reverse wave that cancels out the target reference signal.
  • the noise reduction signal may be played to the user.
  • FIG. 6 is an architectural diagram of an active noise reduction method provided by an embodiment of the present application.
  • the architecture diagram shown in Figure 6 includes the following steps:
  • Step 1 Obtain the target audio information of the user.
  • the target audio signal of the user may refer to an audio signal that the user is interested in, or may refer to an audio signal related to the user's needs.
  • the user's target audio signal may be acquired based on the user's brainwave data.
  • audio signals that the user is currently paying attention to or are interested in may be acquired based on the user's current brainwave data.
  • the external environmental audio signal can be sent to the user more quickly; Noise reduction device.
  • a collaboration device may refer to a device independent of a noise reduction device, for example, a collaboration device may refer to an electronic device with an audio information transfer function; the collaboration device ratio may include a mobile phone, a watch, etc. electronic product.
  • the cooperation device may be one device or multiple devices; the cooperation device may report the location information to the noise reduction device; the timing of reporting the location information may include: reporting the location information when sending the ambient audio signal to the noise reduction device, or Refers to the collaborative device reporting location information when establishing a connection with the noise reduction device for the first time, or the collaborative device can be set to the default location/required location.
  • the user by using the cooperative device to forward the collected ambient audio signals, the user can perceive these signals before the ambient audio signals reach the user; thus, the target of the user's attention can be obtained more efficiently audio information.
  • audio signals that the user is currently paying attention to or are interested in may be acquired based on the user's current brainwave data.
  • the collaborative device may perform the following sub-steps:
  • Sub-step 1 the collaborative device collects ambient audio signals.
  • Sub-step 2 The collaborative device sends an ambient audio signal.
  • the collaboration device can send the collected ambient audio signal d(n) to the noise reduction device; for example, the ambient audio signal can be transmitted by means of radio, such as bluetooth, wifi, 5G, etc.
  • the collaboration device may send the ambient audio signal to the noise reduction device, and before playing the ambient audio signal on the noise reduction device, the energy level of the ambient audio signal may be changed by the noise reduction device.
  • Perform scaling processing so that the playing sound does not affect the user's normal perception; it is used to trigger the user's early perception of the audio information of interest, so as to form this audio perception before the arrival of the direct sound wave d(n), and target based on brainwave data. Presentation of audio information.
  • the collected ambient audio signal may also be scaled and processed by the cooperation device before being forwarded to the noise reduction apparatus.
  • Step 2 Collect ambient audio signals.
  • the noise signal x(n) of the environment where the user is located may be collected by the microphone device in the noise reduction device, and the noise signal x(n) includes the audio signal s(n) that the user is interested in, and the noise signal x(n) relative to the audio signal s( n) Audio signal n(n) that the user does not pay attention to.
  • Step 3 Audio signal filtering processing.
  • the target audio signal is filtered from the noise signal x(n) to obtain the target reference signal x'(n).
  • the noise signal x(n) includes the audio signal s(n) that the user is interested in, and the audio signal n(n) that the user is not concerned about relative to the audio signal s(n); n) After filtering, the obtained target reference signal x'(n) only includes the part of the audio signal n(n) that the user does not pay attention to.
  • Step 4 Perform active noise reduction processing.
  • the environmental audio signal d(n) is the collected audio signal directly to the noise reduction device, that is, d(n) includes both the audio signal that the user pays attention to, and the audio signal that the user does not pay attention to;
  • the audio signal is subjected to active noise reduction processing, so that the noise reduction signal after the noise reduction processing includes the target audio information concerned by the user, so as to achieve better selective listening by the user.
  • FIG. 8 is a schematic diagram of a method for acquiring a target audio signal based on user brainwave data.
  • the method 300 includes steps S310 to S330, and the steps S310 to S330 are described in detail below respectively.
  • the execution body of the method 300 may be a noise reduction device, or may also refer to an brainwave data processing device independent of the noise reduction device, and the brainwave data processing device is used to obtain the user's brainwave data for analysis and obtain the user's brainwave data.
  • the target audio information of interest may be a noise reduction device, or may also refer to an brainwave data processing device independent of the noise reduction device, and the brainwave data processing device is used to obtain the user's brainwave data for analysis and obtain the user's brainwave data.
  • the target audio information of interest may be a noise reduction device, or may also refer to an brainwave data processing device independent of the noise reduction device, and the brainwave data processing device is used to obtain the user's brainwave data for analysis and obtain the user's brainwave data.
  • the target audio information of interest may be a noise reduction device, or may also refer to an brainwave data processing device independent of the noise reduction device, and the brainwave data processing device is used to obtain the user's brainwave data for analysis and obtain the user's brainwave data
  • the user's brainwave data may be acquired by using a brainwave acquisition instrument or other brainwave acquisition sensors.
  • the user's brainwave data may be collected when the user enters an account password to log in to the medical system, which includes brainwave signal values at different time points, or may be real-time data obtained by an brainwave collector.
  • the brainwave data may also be collected under other actions, which is not limited in this embodiment.
  • a large amount of vocal tract motion information when a person speaks can be associated with the user's brainwave data (eg, neural activity data);
  • the motion information may include motion information of the lips, tongue, larynx and jaw.
  • the audio information corresponding to the channel motion information can be obtained by obtaining a pre-trained recurrent neural network through the training data; wherein, the training data may include sample channel motion information and sample audio information, and the recurrent neural network is used to establish the channel operation. The association between information and audio information.
  • the method for obtaining the target audio information of the user shown in FIG. 8 is for illustration, and other methods in the prior art can also be used to obtain the target audio information of the user through brainwave data, or, the downward descending method can also be used.
  • the manner in which the user's target audio information is preconfigured in the noise device.
  • the target audio signal may refer to preconfigured text information or voice information that the user pays attention to; this embodiment of the present application does not make any limitation on this.
  • the method for filtering and processing the audio signal in the above step 3 may refer to FIG. 9 .
  • an adaptive filtering architecture can be used for filtering, and the obtained target audio signal of the user is removed from the noise signal x(n) obtained by direct acquisition, so as to obtain the target reference signal x'(n).
  • FIG. 9 the architecture shown in FIG. 9 is similar in principle to the ANC noise reduction algorithm model.
  • the error signal e(n) of the adaptive filter is:
  • x(n) represents the signal to be filtered, that is, the initial reference signal collected
  • x'(n) represents the signal output after filtering, that is, the target reference signal
  • s'(n) represents the user's target audio signal, that is, the user's attention audio signal.
  • the target audio signal s'(n) of the user's attention in the noise signal x(n) can be filtered out, thereby obtaining the target reference signal x'(n) that does not include the target audio signal of the user's attention.
  • the method for acquiring target audio information based on brainwave data in FIG. 8 may be to acquire the user's target audio information by collecting the user's brainwave data in real time and analyzing it.
  • obtaining the user's target audio information through brainwave data may not meet some real-time service requirements; therefore, in the embodiment of the present application, a prediction model may also be used to obtain the user's target audio information; the prediction model uses The target audio information of the user in the future is predicted based on the brain wave data of the user at the historical moment or the current moment and the target audio information of the user.
  • the prediction model can use a recurrent neural network, and the output of the recurrent neural network not only depends on the current input data, but also on the state of the previous moment; through the introduction of the time series feedback mechanism, the context information can be effectively used, in speech, text, etc.
  • the analysis of timing signals is of great significance.
  • the input of a recurrent neural network can be composed of two parts, namely the hidden layer state information of the recurrent neural network at the previous moment and the current input information.
  • the calculation formula is as follows:
  • W, U, b are the weight parameters of the model, where h t represents the hidden layer state at time t; x t represents the input at time t; f represents the nonlinear activation function, for example, tanh function, sigmoid function, ReLu function, etc. can be used as the activation function.
  • V and c are parameters, represents the output of the recurrent neural network at time t.
  • the output dimension of a recurrent neural network can be the same size as the number of words in the vocabulary.
  • cross-entropy can be used as a loss function for training to obtain the required language model, and the prediction model can predict the target audio information of the user at the next moment based on the target audio information of the user at the historical moment.
  • voice information can be converted into text information first; specifically, it includes:
  • Step 1 Convert the user's target audio signal s'(n) obtained at the current moment into a voice text text'(n) based on speech recognition;
  • Step 2 Use text'(n) as the input of the recurrent neural network, and then obtain the output text"(n) based on the pre-trained recurrent neural network, and text"(n) can be used to represent the predicted current moment at the next moment.
  • User's target audio information
  • Step 3 the text "(n) is subjected to speech synthesis to obtain the target audio information s"(n) at the next moment of the predicted current moment;
  • the collected noise signal x(n) can be removed by the method shown in FIG. 9 according to the target audio information of the user based on the prediction after supplementation to obtain the target reference signal x'(n).
  • a prediction model is introduced when acquiring the user's target audio information based on brainwave data, that is, acquiring the audio information that the user is concerned about; thus, the time for acquiring the user's target audio information can be further reduced.
  • the wave data and the predicted target audio information of the user can better understand the audio information of the user's attention, so that the audio information of the user's attention can be better retained in the noise reduction signal, so that the user can choose to listen better.
  • the target reference signal is obtained by acquiring the target audio signal that the user pays attention to; the target reference signal does not include the target audio signal of the user; according to the target reference signal and the ambient audio signal, the noise reduction signal is obtained, and the noise reduction signal is obtained.
  • the signal is used to offset the target reference signal; the noise reduction signal after the active noise reduction process still includes the target audio signal of the user's attention, so that the active noise reduction method can meet the individual needs of the user.
  • the active noise reduction method provided by the embodiments of the present application is described in detail above with reference to FIGS. 1 to 9 ; the device embodiments of the present application will be described in detail below with reference to FIGS. 10 to 12 . It should be understood that the active noise reduction apparatus in the embodiments of the present application can perform various active noise reduction methods in the foregoing embodiments of the present application, that is, for the specific working processes of the following various products, reference may be made to the corresponding processes in the foregoing method embodiments.
  • FIG. 10 is a schematic block diagram of an active noise reduction device provided by the present application.
  • the active noise reduction apparatus 400 may perform the active noise reduction methods shown in FIGS. 5 to 9 .
  • the active noise reduction device 400 includes: an acquisition module 410 and a processing module 420 .
  • the acquisition module 410 is used to acquire the target audio signal, the environmental audio signal and the initial reference signal of the user, wherein the target audio signal is used to represent the audio information concerned by the user, and the environmental audio signal is used to represent the Audio information of the environment where the user is located, the initial reference signal is used to perform active noise reduction processing on the environmental audio signal;
  • the processing module 420 is configured to obtain a target reference signal according to the target audio signal and the initial reference signal, The target reference signal does not include the target audio signal; the noise reduction signal is obtained according to the target reference signal and the ambient audio signal, and the noise reduction signal is used to cancel the target reference signal.
  • the obtaining module 410 is specifically configured to:
  • the processing module is specifically used for:
  • the target audio signal of the user is acquired according to the brain wave data of the user.
  • the brainwave data is used to reflect the neuron state of the environmental audio signal acquired by the user, and the target audio signal refers to the user's neuronal state in the environmental audio signal.
  • An audio signal whose meta state satisfies a preset condition.
  • the target audio signal refers to an audio signal preconfigured according to the behavior log of the user.
  • processing module 420 is specifically configured to:
  • the vocal tract motion information is used to represent the motion information of the vocal tract occlusal part when the user speaks;
  • the target audio signal is obtained according to the channel motion information.
  • the target audio signal is the target audio signal at the current moment
  • the processing module 420 is further configured to:
  • the target audio signal at the next moment at the current moment is predicted according to the target audio signal at the current moment.
  • processing module 420 is specifically configured to:
  • the initial reference signal is filtered according to the target audio signal to obtain the target reference signal.
  • processing module 420 is specifically configured to:
  • the initial reference signal is filtered by using an adaptive filter according to the target audio signal.
  • the obtaining module 410 is specifically configured to:
  • the ambient audio signal is received from a cooperating device for acquiring the ambient audio signal from a sound source.
  • the active noise reduction apparatus 400 may be used to perform all or part of the operations of the method shown in any one of FIGS. 5 to 9 .
  • the acquiring module can be used to perform all or part of the operations in S210 and S310; the processing module can be used to perform all or part of the operations in S220, S230, S320 and S330.
  • the acquisition module 410 may refer to a communication interface or a transceiver in the active noise reduction device; for example, the acquisition module 410 may refer to a microphone or an interface circuit in the active noise reduction device.
  • the processing module 420 may be a processor or chip with computing capability in any active noise reduction device.
  • module here can be implemented in the form of software and/or hardware, which is not specifically limited.
  • a “module” may be a software program, a hardware circuit, or a combination of the two that implement the above-mentioned functions.
  • the hardware circuits may include application specific integrated circuits (ASICs), electronic circuits, processors for executing one or more software or firmware programs (eg, shared processors, proprietary processors, or group processors) etc.) and memory, merge logic and/or other suitable components to support the described functions.
  • ASICs application specific integrated circuits
  • processors for executing one or more software or firmware programs eg, shared processors, proprietary processors, or group processors
  • the units of each example described in the embodiments of the present application can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.
  • FIG. 11 is a schematic diagram of an active noise reduction device.
  • the acquisition module 410 shown in FIG. 10 may include the brainwave data acquisition module 510 and the ambient audio signal acquisition module 520 shown in FIG. 11 ; the processing module 420 may include a filtering processing module 530 and a noise reduction processing module 540 .
  • the brain wave data acquisition module 510 is used for acquiring the user's brain wave data; and obtaining the user's target audio signal according to the brain wave data.
  • the brain wave data acquisition module 510 may be used to perform S310 to S330.
  • the ambient audio signal acquisition module 520 is configured to acquire the ambient audio signal where the user is located; for example, the ambient audio signal can be acquired through the microphone of the active noise reduction device.
  • the ambient audio signal acquisition module 520 may be configured to perform S210.
  • the filtering processing module 530 is configured to perform filtering processing on the target audio signal of the user from the initial reference signal to obtain the target reference signal.
  • the ambient audio signal acquisition module 520 may be configured to perform S220.
  • the noise reduction processing module 540 is configured to perform active noise reduction processing on the environmental audio signal according to the target reference signal to obtain a noise reduction signal; wherein the noise reduction signal includes the user's target audio signal, that is, the user can still hear the audio information that the user is concerned about. .
  • the noise reduction processing module 540 may be used to perform S230.
  • FIG. 12 is a schematic diagram of a hardware structure of an active noise reduction device provided by an embodiment of the present application.
  • the active noise reduction apparatus 600 shown in FIG. 12 includes a memory 610 , a processor 620 , a communication interface 630 and a bus 640 .
  • the memory 610 , the processor 620 , and the communication interface 630 are connected to each other through the bus 640 for communication.
  • the memory 610 may be a read only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM).
  • the memory 610 may store a program, and when the program stored in the memory 610 is executed by the processor 620, the processor 620 is configured to execute each step of the active noise reduction method of the embodiment of the present application; each step.
  • the active noise reduction device shown in the embodiment of the present application may be an active noise reduction earphone, or a chip configured in the active noise reduction earphone; or, the active noise reduction device shown in the embodiment of the present application may be a vehicle
  • the headrest device may also be a chip arranged in the vehicle headrest device.
  • the active noise reduction device may be a device with an active noise reduction function, for example, may include any device known in the current technology; or, the active noise reduction device may also refer to a chip with an active noise reduction function.
  • the active noise reduction apparatus may include a memory and a processor; the memory may be used to store program codes, and the processor may be used to call the program codes stored in the memory to implement corresponding functions of the computing device.
  • the processor and memory included in the computing device may be implemented by a chip, which is not specifically limited here.
  • the memory may be used to store relevant program instructions of the active noise reduction method provided in the embodiments of the present application
  • the processor may be used to call the relevant program instructions of the active noise reduction method stored in the memory.
  • the processor 620 can be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for executing relevant programs to achieve The active noise reduction method of the method embodiment of the present application.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • the processor 620 may also be an integrated circuit chip with signal processing capability.
  • each step of the active noise reduction method of the present application may be completed by an integrated logic circuit of hardware in the processor 920 or instructions in the form of software.
  • the above-mentioned processor 620 can also be a general-purpose processor, a digital signal processor (digital signal processing, DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, Discrete gate or transistor logic devices, discrete hardware components.
  • DSP digital signal processing
  • ASIC application-specific integrated circuit
  • FPGA field programmable gate array
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the steps of the methods disclosed in conjunction with the embodiments of the present application may be directly embodied as executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other storage media mature in the art.
  • the storage medium is located in the memory 610, and the processor 620 reads the information in the memory 610, and completes the functions required by the modules included in the active noise reduction device shown in FIG. 10 or FIG. 11 in the implementation of this application in combination with its hardware, or, The active noise reduction method shown in FIG. 5 to FIG. 9 in the method embodiment of the present application is performed.
  • the communication interface 630 uses a transceiver such as, but not limited to, a transceiver to enable communication between the active noise cancelling device 600 and other devices or communication networks.
  • Bus 640 may include pathways for communicating information between various components of active noise cancelling device 600 (eg, memory 610, processor 620, communication interface 630).
  • the active noise reduction apparatus 600 only shows a memory, a processor, and a communication interface, in the specific implementation process, those skilled in the art should understand that the active noise reduction apparatus 600 may also include necessary components for normal operation. of other devices. Meanwhile, those skilled in the art should understand that the above-mentioned active noise reduction apparatus 600 may further include hardware devices that implement other additional functions according to specific needs. In addition, those skilled in the art should understand that the above-mentioned active noise reduction apparatus 600 may also only include the necessary components for implementing the embodiments of the present application, rather than all the components shown in FIG. 12 .
  • an embodiment of the present application further provides an active noise reduction earphone, which can perform the active noise reduction method in the foregoing method embodiments.
  • the embodiments of the present application further provide a vehicle headrest device, which can perform the active noise reduction method in the above method embodiments.
  • the vehicle headrest device may refer to an active noise reduction headrest of a car seat, a through hole may be opened on the surface of the headrest body, one or more noise reduction speakers may be installed in the through hole, and the noise reduction speaker may be used for A noise reduction signal is output, and the noise reduction signal may be a noise reduction signal obtained according to the active noise reduction method in the embodiment of the present application.
  • one or more noise-cancelling speakers may be installed in the headrest near the user's ear, so that the user can receive noise-cancelling signals output from the noise-cancelling speakers.
  • one or more microphones may also be installed on the headrest body, and the microphones may be used to collect ambient audio signals or initial reference signals; wherein, the microphones used to collect ambient audio signals or initial reference signals may be the same Microphone, can also refer to different microphones.
  • the ambient audio signal may be the audio information in the cockpit perceived by the user when the active noise canceling headrest is used and the noise canceling function is not turned on;
  • the initial reference signal may refer to the ambient audio for the car
  • the signal is a noise signal subjected to active noise reduction processing;
  • the initial reference signal may include one or more of an engine noise signal, a wind noise signal, and a road noise signal of an automobile.
  • a microphone for collecting ambient audio signals or initial reference signals can be installed anywhere in the cabin.
  • the active noise reduction headrest of the car seat may refer to the active noise reduction headrest installed on the driver's seat of the car; or, it may also refer to the headrest installed on the passenger seat of the car. Active noise reduction headrest; the embodiment of the present application does not make any limitation on the specific installation position of the active noise reduction headrest.
  • the embodiments of the present application further provide an automobile, including the active noise reduction device in the above method embodiments.
  • the active noise-cancelling headrest of the above-described embodiments may be included in a car.
  • an embodiment of the present application further provides an active noise reduction system, which can execute the active noise reduction method in the foregoing method embodiments.
  • an embodiment of the present application further provides a chip, where the chip includes a transceiver unit and a processing unit.
  • the transceiver unit may be an input/output circuit or a communication interface;
  • the processing unit may be a processor, a microprocessor or an integrated circuit integrated on the chip; and the chip may execute the active noise reduction method in the above method embodiments.
  • an embodiment of the present application further provides a computer-readable storage medium, on which an instruction is stored, and when the instruction is executed, the active noise reduction method in the foregoing method embodiment is executed.
  • an embodiment of the present application further provides a computer program product including an instruction, when the instruction is executed, the active noise reduction method in the foregoing method embodiment is executed.
  • the processor in the embodiment of the present application may be a central processing unit (central processing unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (digital signal processors, DSP), application-specific integrated circuits (application specific integrated circuit, ASIC), off-the-shelf programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the memory in the embodiments of the present application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically programmable Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • Volatile memory may be random access memory (RAM), which acts as an external cache.
  • RAM random access memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • DDR SDRAM double data rate synchronous dynamic random access memory
  • enhanced SDRAM enhanced synchronous dynamic random access memory
  • SLDRAM synchronous connection dynamic random access memory Fetch memory
  • direct memory bus random access memory direct rambus RAM, DR RAM
  • the above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination.
  • the above-described embodiments may be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated.
  • the computer may be a general purpose computer, special purpose computer, computer network, or other programmable device.
  • the computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server or data center by wire (eg, infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that a computer can access, or a data storage device such as a server, a data center, or the like containing one or more sets of available media.
  • the usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media.
  • the semiconductor medium may be a solid state drive.
  • At least one means one or more, and “plurality” means two or more.
  • At least one item(s) below” or similar expressions thereof refer to any combination of these items, including any combination of single item(s) or plural items(s).
  • at least one item (a) of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple .
  • the size of the sequence numbers of the above-mentioned processes does not mean the sequence of execution, and the execution sequence of each process should be determined by its functions and internal logic, and should not be dealt with in the embodiments of the present application. implementation constitutes any limitation.
  • the disclosed system, apparatus and method may be implemented in other manners.
  • the apparatus embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium.
  • the technical solution of the present application can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Quality & Reliability (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Soundproofing, Sound Blocking, And Sound Damping (AREA)
  • Circuit For Audible Band Transducer (AREA)

Abstract

一种主动降噪方法、主动降噪装置以及主动降噪系统，应用主动降噪方法的主动降噪耳机、车载头靠装置，以及包含主动降噪装置的汽车。该主动降噪方法包括：获取用户的目标音频信号、环境音频信号以及初始参考信号，其中，目标音频信号用于表示用户关注的音频信息，环境音频信号用于表示用户所处环境的音频信息，初始参考信号用于对环境音频信号进行主动降噪处理（S210）；根据目标音频信号与初始参考信号，得到目标参考信号，目标参考信号中不包括目标音频信号（S220）；根据目标参考信号与环境音频信号，得到降噪信号，降噪信号用于抵消目标参考信号（S230）。该方法能够根据用户的需求进行选择性的降噪处理，从而使得主动降噪方法满足用户的个性化需求。

Description

主动降噪方法、主动降噪装置以及主动降噪系统 技术领域
本申请涉及主动降噪领域，尤其涉及一种主动降噪方法、主动降噪装置以及主动降噪系统。
背景技术
主动降噪(active noise cancellation，ANC)是基于声波叠加原理，通过声波互相抵消来实现噪声去除。
目前，在主动降噪系统中任何外部的音频信号均被进行抑制；但是，通常情况下用户依然希望对自身关注的音频信息进行有效的感知；比如，用户感兴趣的音频信号或者与需求相关的实时音频信号。
因此，在主动降噪时如何避免将任何外部的声音信号均降噪，使得主动降噪方法满足用户的个性化需求成为一个亟需解决的问题。
发明内容
本申请提供一种主动降噪方法以及主动降噪方法、主动降噪装置以及主动降噪系统，使得主动降噪处理后的降噪信号中依然包括用户关注的目标音频信号，能够满足用户的个性化需求。
第一方面，提供了一种主动降噪方法，包括：
获取用户的目标音频信号、环境音频信号以及初始参考信号，其中，所述目标音频信号用于表示所述用户关注的音频信息，所述环境音频信号用于表示所述用户所处环境的音频信息，所述初始参考信号用于表示对所述环境音频信号进行主动降噪处理；根据所述目标音频信号与所述初始参考信号，得到目标参考信号，所述目标参考信号中不包括所述目标音频信号；所述根据所述目标参考信号与所述环境音频信号，得到降噪信号，所述降噪信号用于抵消所述目标参考信号。
应理解，对于不同的用户，目标音频信号可以不同。目标音频信号可以是指用户关注的音频信号；比如，目标音频信号可以是指环境音频信号中用户感兴趣的音频信号，或者，环境音频信号中与用户的需求相关的音频信号。获取用户的目标音频信号的方式可以包括但不限于：通过获取的用户的脑波数据所反映的用户的神经元状态得到用户的神经元状态满足预设条件时对应的音频信息，即为该用户关注的音频信息；或者，可以通过获取用户的行为日志获取用户关注的音频信息；其中，用户的行为日志可以包括用户的兴趣爱好、用户的下载历史信息、用户的浏览记录信息等。
需要说明的是，环境音频信号可以是指用户在使用降噪装置并未开启降噪功能时感知到的所在环境中的音频信息；初始参考信号可以是指通过降噪装置中的器件采集的音频信号；比如，可以是指通过降噪装置的麦克风采集的音频信号。应理解，环境音频信号与初 始参考信号中包括的音频信息可以存在差异。其中，任何辅助进行降噪功能的设备或器件都可以认为是降噪装置的一部分。
在本申请中，通过获取用户关注的目标音频信号，从而得到目标参考信号；其中，目标参考信号中不包括用户的目标音频信号；根据目标参考信号与环境音频信号，得到降噪信号，降噪信号用于抵消目标参考信号；使得主动降噪处理后的降噪信号中依然包括用户关注的目标音频信息，从而使得主动降噪方法能够满足用户的个性化需求。
结合第一方面，在第一方面的某些实现方式中，所述获取用户的目标音频信号，包括：
获取所述用户的脑波数据；
根据所述用户的脑波数据获取所述用户的目标音频信号。
在本申请的实施例中，可以根据获取的用户对环境音频信号的脑波数据，从而获取用户关注的目标音频信号。
应理解，上述脑波数据又可以称为脑电波数据，是指通过使用电生理指标记录大脑活动的方法获取的用户的大脑的数据。用户的大脑在活动时，大量神经元同步发生的突触后电位经总和后形成脑电波，它记录大脑活动时的电波变化，是脑神经细胞的电生理活动在大脑皮层或头皮表面的总体反映。对于不同用户而言，脑波数据可以是不同的。
结合第一方面，在第一方面的某些实现方式中，所述脑波数据用于反映所述用户在获取到所述环境音频信号时的神经元状态，所述目标音频信号是指所述环境音频信号中所述用户的神经元状态满足预设条件的音频信号。
例如，预设条件可以包括用户的神经元状态的波动范围满足预设阈值，或者，其它预先配置的条件；比如，使得神经元状态波动较大的音频信号可以为目标音频信号。
结合第一方面，在第一方面的某些实现方式中，所述目标音频信号是指根据所述用户的行为日志预先配置的音频信号。
示例性地，用户的行为日志兴趣爱好、用户的下载历史信息、用户的浏览记录信息等。
在一种可能的实现方式中，可以根据预先配置的用户信息获取用户的目标音频信号。比如，目标音频信号可以是指预先配置的用户关注的文本信息，或者语音信息。
结合第一方面，在第一方面的某些实现方式中，所述根据所述用户的脑波数据获取所述用户的目标音频信号，包括：
将所述用户的脑波数据转换为声道运动信息，所述声道运动信息用于表示所述用户讲话时声道咬合部位的运动信息；
根据所述声道运动信息得到所述目标音频信号。
需要说明的是，为了实现脑波数据到声道运动信息的转化和映射，可以将人说话时的大量声道运动信息与用户的脑波数据(例如，神经活动数据相关联)；其中，声道运行信息可以包括嘴唇、舌头、喉和下颌的运动信息；可以通过训练数据得到预先训练的循环神经网络得到声道运动信息对应的音频信息；其中，训练数据可以包括样本声道运动信息与样本音频信息，循环神经网络用于建立声道运行信息与音频信息之间的关联关系。
结合第一方面，在第一方面的某些实现方式中，所述目标音频信号为当前时刻的目标音频信号，还包括：
根据所述当前时刻的目标音频信号预测所述当前时刻的下一时刻的目标音频信号。
应理解，当前时刻与当前时刻的下一时刻可以是连续的时刻，或者，也可以是指不连 续的时刻；比如，当前时刻的下一时刻可以与当前时刻相差周期性时间间隔，该时间间隔的时间单位可以包括毫秒或者微秒。
在本申请的实施例中，由于通过脑波数据获取用户的目标音频信号可能无法满足部分实时性业务需求；因此，还可以采用预测模型获取用户的目标音频信号；预测模型用于基于历史时刻或者当前时刻用户的脑波数据以及用户的目标音频信号预测未来时刻用户的目标音频信号。
结合第一方面，在第一方面的某些实现方式中，所述根据所述目标音频信号与初始参考信号，得到目标参考信号，包括：
根据所述目标音频信号对所述初始参考信号进行滤波处理，得到所述目标参考信号。
结合第一方面，在第一方面的某些实现方式中，所述根据所述目标音频信号对所述初始参考信号进行滤波处理，包括：
根据所述目标音频信号通过采用滤波器对所述初始参考信号进行滤波处理。
在一种可能的实现方式中，上述滤波器可以包括自适应滤波器，或者其它滤波器。
结合第一方面，在第一方面的某些实现方式中，所述获取环境音频信号，包括：
接收来自协作设备所述环境音频信号，所述协作设备用于从声源处获取所述环境音频信号。
在本申请的实施例中，通过采用协作设备对采集的环境音频信号进行转发可以使得在环境音频信号在到达用户之前，用户就可以感知到这些信号；从而能够更高效的获取用户的关注的目标音频信号。
在一种可能的实现方式中，协作设备可以将环境音频信号发送至降噪装置，在降噪装置上在播放环境音频信号之前，可以通过降噪装置对环境音频信号的能级进行缩放处理，使得播放声音不影响用户的正常感知；用于触发用户对关注的音频信息的提前感知，从而在直达声波到来之前形成这种音频感知，并基于脑波数据进行目标音频信息的呈现。
在一种可能的实现方式中，也可以由协作设备对采集的环境音频信号进行缩放处理后再转发至降噪装置。
第二方面，提供了一种主动降噪装置，包括：
获取模块，用于获取用户的目标音频信号、环境音频信号以及初始参考信号，其中，所述目标音频信号用于表示所述用户关注的音频信息，所述环境音频信号用于表示所述用户所处环境的音频信息，所述初始参考信号用于对所述环境音频信号进行主动降噪处理；处理模块，用于根据所述目标音频信号与所述初始参考信号，得到目标参考信号，所述目标参考信号中不包括所述目标音频信号；所述根据所述目标参考信号与所述环境音频信号，得到降噪信号，所述降噪信号用于抵消所述目标参考信号。
结合第二方面，在第二方面的某些实现方式中，所述获取模块具体用于：
获取所述用户的脑波数据；
所述处理模块具体用于：
根据所述用户的脑波数据获取所述用户的目标音频信号。
结合第二方面，在第二方面的某些实现方式中，所述脑波数据用于反映所述用户获取所述环境音频信号时的神经元状态，所述目标音频信号是指所述环境音频信号中所述用户的神经元状态满足预设条件的音频信号。
结合第二方面，在第二方面的某些实现方式中，所述目标音频信号是指根据所述用户的行为日志预先配置的音频信号。
结合第二方面，在第二方面的某些实现方式中，所述处理模块具体用于：
将所述用户的脑波数据转换为声道运动信息，所述声道运动信息用于表示所述用户讲话时声道咬合部位的运动信息；
根据所述声道运动信息得到所述目标音频信号。
结合第二方面，在第二方面的某些实现方式中，所述目标音频信号为当前时刻的目标音频信号，所述处理模块还用于：
根据所述当前时刻的目标音频信号预测所述当前时刻的下一时刻的目标音频信号。
结合第二方面，在第二方面的某些实现方式中，所述处理模块具体用于：
根据所述目标音频信号对所述初始参考信号进行滤波处理，得到所述目标参考信号。
结合第二方面，在第二方面的某些实现方式中，所述处理模块具体用于：
根据所述目标音频信号通过采用自适应滤波器对所述初始参考信号进行滤波处理。
结合第二方面，在第二方面的某些实现方式中，所述获取模块具体用于：
接收来自协作设备的所述环境音频信号，所述协作设备用于从声源处获取所述环境音频信号。
第三方面，提供了一种主动降噪装置，包括存储器，用于存储程序；处理器，用于执行该存储器存储的程序，当该存储器存储的程序被执行时，该处理器用于执行：获取用户的目标音频信号、环境音频信号以及初始参考信号，其中，所述目标音频信号用于表示所述用户关注的音频信息，所述环境音频信号用于表示所述用户所处环境的音频信息，所述初始参考信号用于对所述环境音频信号进行主动降噪处理；根据所述目标音频信号与所述初始参考信号，得到目标参考信号，所述目标参考信号中不包括所述目标音频信号；所述根据所述目标参考信号与所述环境音频信号，得到降噪信号，所述降噪信号用于抵消所述目标参考信号。
在一种可能的实现方式中，上述主动降噪装置中包括处理器还用于执行第一方面以及第一方面中的任意一种实现方式中的主动降噪方法。
应理解，在上述第一方面中对相关内容的扩展、限定、解释和说明也适用于第三方面中相同的内容。
第四方面，提供了一种主动降噪耳机，用于执行第一方面以及第一方面中的任意一种实现方式中的主动降噪方法。
应理解，在上述第一方面中对相关内容的扩展、限定、解释和说明也适用于第四方面中相同的内容。
第五方面，提供了一种车载头靠装置，用于执行第一方面以及第一方面中的任意一种实现方式中的主动降噪方法。
应理解，在上述第一方面中对相关内容的扩展、限定、解释和说明也适用于第五方面中相同的内容。
第六方面，提供了一种汽车，包括执行第二方面以及第二方面中的任意一种实现方式中的主动降噪装置。
应理解，在上述第一方面中对相关内容的扩展、限定、解释和说明也适用于第六方面 中相同的内容。
第七方面，提供了一种主动降噪系统，用于执行第二方面以及第二方面中的任意一种实现方式中的主动降噪装置。
应理解，在上述第一方面中对相关内容的扩展、限定、解释和说明也适用于第七方面中相同的内容。
第八方面，提供了一种包含指令的计算机程序产品，当该计算机程序产品在计算机上运行时，使得计算机执行上述第一方面以及第一方面中的任意一种实现方式中的主动降噪方法。
第九方面，提供了一种芯片，所述芯片包括处理器与数据接口，所述处理器通过所述数据接口读取存储器上存储的指令，执行上述第一方面以及第一方面中的任意一种实现方式中的主动降噪方法。
可选地，作为一种实现方式，所述芯片还可以包括存储器，所述存储器中存储有指令，所述处理器用于执行所述存储器上存储的指令，当所述指令被执行时，所述处理器用于执行上述第一方面以及第一方面中的任意一种实现方式中的主动降噪方法。
附图说明
图1为主动降噪系统的示意性框图；
图2为主动降噪系统的原理示意图；
图3为降噪信号与噪声信号叠加相消的示意图；
图4为主动降噪系统的基本原理的示意图；
图5是本申请实施例提供的主动降噪方法的示意性流程图；
图6是本申请实施例提供的主动降噪方法的架构图；
图7是本申请实施例提供的通过协作设备获取环境音频信号的示意图；
图8为基于用户脑波数据获取目标音频信号的方法的示意图；
图9为对音频信号进行滤波处理的示意图；
图10是本申请实施例提供的主动降噪装置的示意性框图；
图11是本申请实施例提供的主动降噪装置的示意性框图；
图12是本申请实施例提供的主动降噪装置的硬件结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述；显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
首先对本申请实施例中涉及的概念进行简单的说明。
主动降噪(active noise cancellation，ANC)是一种基于声波叠加原理，通过声波互相抵消来实现噪声去除的技术；主动降噪系统可以包括前馈型与反馈型。
作为示例，下面先结合图1、图2、图3与图4详细描述主动降噪系统的组成与降噪 原理。
示例性地，如图1所示，主动降噪系统100通常可以包括控制器110、扬声器(speaker)120、误差传感器(error mic)130、参考传感器(ref mic)140。
参见图2，对图1所示主动降噪系统的工作原理与工作流程进行详细的描述。
步骤①：误差传感器130采集误差信号e(n)，并将误差信号e(n)传递到控制器110。
其中，误差信号e(n)用于表示图2所示静区内的声场特征；例如，该声场特性包括声压特性。关于静区的概念下文将描述，这里暂不详述。
误差传感器130可以包括声学传感器、光学传感器或者其它传感器；误差传感器130可以通过采集振动获取音频信息。例如，图2与图3所示，误差传感器130可以包括麦克风。
步骤②：参考传感器140采集噪声信号x(n)，并将噪声信号x(n)传递到控制器110。
示例性地，噪声信号x(n)可以是指本申请中用于对环境音频信号进行主动降噪处理的初始参考信号，即在本申请的实施例中可以将参考传感器采集到的噪声信号x(n)用于对环境音频信号进行主动降噪处理的初始参考信号。
应理解，参考传感器140采集的噪声信号x(n)是环境噪声信号。环境噪声信号通常是由不期望的噪声源发出的，如图2中所示。
参考传感器140通常为声学传感器。如图2与图3所示，参考传感器140为麦克风。
步骤③：控制器110基于误差信号e(n)计算误差代价函数，并基于误差代价函数最小化原则，基于噪声信号x(n)预测扬声器120输出的降噪信号y(n)。
降噪信号y(n)用于抵消噪声信号x(n)。理想情况是，降噪信号y(n)是噪声信号x(n)的反相信号，即降噪信号y(n)与噪声信号x(n)相加后为零。降噪信号y(n)也可称为抗噪信号。
例如，控制器110可以为自适应滤波器。
步骤④：扬声器120根据控制器110的控制，输出降噪信号y(n)；比如，控制器110通过计算到达静区后叠加后的信号e(n)的误差代价函数；并基于误差代价函数最小原则控制扬声器120的输出信号。或者，扬声器120根据控制器110产生的用于抵消噪声的反向信号控制扬声器输出的降噪信号y(n)。如图2所示，噪声信号x(n)和降噪信号y(n)分别经过初级通路和次级通路到达静区。
如图3所示，误差传感器130采集的是，噪声信号x(n)和降噪信号y(n)分别经过初级通路和次级通路到达静区后叠加后的声音信号，该声音信号被称为误差信号e(n)。误差传感器130采集的噪声信号e(n)也可以描述为，是经降噪处理后的残差噪声。
控制器110预测扬声器120输出的降噪信号y(n)的目标是，使得噪声信号x(n)和降噪信号y(n)分别经过初级通路和次级通路到达静区后叠加后的信号e(n)的误差代价函数最小。
例如，若将噪声源视为初级声源，可以将扬声器120称为次级声源，如图2所示。
应理解，噪声信号与降噪信号在不同位置的叠加效果不一定相同。假设，误差传感器采集A点的误差信号，该误差信号可以表征噪声信号与降噪信号在A点的叠加效果，但不一定可以表征噪声信号与降噪信号在A点之外其他位置的叠加效果。为了表达主动降噪的误差信号代表的是哪个区域的主动降噪效果，静区(quiet zone)的概念被提出来，表示误差传感器所采集的误差信号所在的区域或空间。也就是说，误差传感器采集哪里的信号， 哪里就是静区。例如，在图2中，静区表示误差传感器130采集的误差信号e(n)所在的区域。
还应理解到，控制器110预测扬声器120输出的降噪信号y(n)的目标是，使得降噪信号y(n)与噪声信号x(n)分别到达静区后叠加后的信号e(n)的误差代价函数最小。也就是说，主动降噪系统是以实现静区内的主动降噪效果为目标。
示例性地，图4为主动降噪系统的基本原理的示意图。
其中，上述主动降噪系统可以是指综合式主动降噪系统，综合式主动降噪系统是前馈式与反馈式的组合。d可以是指信号源信号，包括有用信号S与噪声干扰n 0之和；x可以是指参考信号，参考信号可以为x＝n 1，n 1是指与噪声干扰n 0相关的信号。假设S、n 0、n 1是零均值的平稳随机过程，且满足S与n 0、n 1互不相关，滤波器的输出信号y＝n 2，n 2可以表示噪声n 1的滤波信号，则整个系统的输出为：
Z＝d-y＝S+n 0-y；
等式两边平方得：
Z 2＝d 2-y 2＝S 2+(n 0-y) 2+2S(n 0-y)；
等式两边取期望值得：
Figure PCTCN2020106681-appb-000001
其中，Ε[Z 2]可以表示信号的功率；由上式可以看出，要使得系统输出Z最大程度地接近信号S，就要使得Ε[(n 0-y) 2]取最小值；而Z-S＝n 0-y，在理想情况下，y＝n 0，则Z＝S；输出信号Z的噪声完全被抵消，只保留有用信号S。
因此，可以看出主动降噪系统的本质是：将d(n)中包含x(n)或者与x(n)相关的信号去除，如果x(n)内容与d(n)基本一致，则d(n)信号就被抵消。。
在主动降噪系统中，是基于参考传感器采集的噪声信号x(n)基本对环境音频信号d(n)又可以称为信号源信号进行降噪处理，从而降噪处理后的降噪信号向用户播放。目前的主动降噪系统中，不会对采集的环境中的噪声信号x(n)进行区分，即对x(n)最大程度的实现全面降噪，任何外部的声音信号均被抑制掉。但是，在日常生活中，用户依然希望对自身的关注音频信息进行有效的感知，比如，用户感兴趣的音频信号或者与需求相关的实时音频信号；因此，当前的主动降噪方法无法满足用户的个性化需求，即无法根据用户的需求进行选择性的降噪处理。
有鉴于此，本申请提出了一种主动降噪方法以及主动降噪装置，通过获取用户关注的目标音频信号，从而得到目标参考信号；其中，目标参考信号中不包括用户的目标音频信号；根据目标参考信号与环境音频信号，得到降噪信号，降噪信号用于抵消目标参考信号；使得主动降噪处理后的降噪信号中依然包括用户关注的目标音频信号，从而使得主动降噪方法能够满足用户的个性化需求。
下面将结合附图5至图9，对本申请中的技术方案进行详细描述。
首先，需要说明的是本申请的主动降噪方法可以应用于降噪装置中，其中，降噪装置是指具有降噪需求的装置；比如，降噪装置可以包括降噪播放装置以及车载场景中的用户头靠装置。
示例性地，降噪播放装置可以包括主动降噪耳机、主动降噪式头带设备等；车载场景中的用户头靠装置可以是指汽车中座椅头靠处用于进行声音播放的装置。
图5为本申请实施例提供的主动降噪方法的示意性流程图。例如，方法200的执行主体为上述降噪装置中的任意一种。方法200包括步骤S210至步骤S230，下面分别对步骤S210至步骤S230进行详细的描述。
S210、获取用户的目标音频信号、环境音频信号以及初始参考信号。
其中，目标音频信号用于表示所述用户关注的音频信息，环境音频信号用于表示用户所处环境的音频信息，初始参考信号用于对环境音频信号进行主动降噪处理。
应理解，目标音频信号可以是指用户关注的音频信号；比如，目标音频信号可以是指环境音频信号中用户感兴趣的音频信号，或者，环境音频信号中与用户的需求相关的音频信号。比如，目标音频信号可以是通过获取用户的脑波数据，根据脑波数据反映用户的神经元状态可以确定用户关注的目标音频信息；或者，目标音频信号也可以是根据用户的行为日志预先配置的音频信息；用户的行为日志可以包括用户的兴趣爱好、用户的下载历史信息等。
需要说明的是，环境音频信号可以是指用户在使用降噪装置并未开启降噪功能时感知到的所在环境中的音频信息；初始参考信号可以是指通过降噪装置中的器件采集的音频信号；比如，可以是指通过降噪装置的麦克风采集的音频信号。应理解，环境音频信号与初始参考信号中包括的音频信息可以存在差异。其中，任何辅助进行降噪功能的设备或器件都可以认为是降噪装置的一部分。
可选地，在一种可能的实现方式中，获取用户的目标音频信号，包括：获取所述用户的脑波数据；根据所述用户的脑波数据获取所述用户的目标音频信号。
应理解，上述脑波数据又可以称为脑电波数据，是指通过使用电生理指标记录大脑活动的方法获取的用户的大脑的数据。用户的大脑在活动时，大量神经元同步发生的突触后电位经总和后形成脑电波，它记录大脑活动时的电波变化，是脑神经细胞的电生理活动在大脑皮层或头皮表面的总体反映。对于不同用户而言，脑波数据可以是不同的。
可选地，在一种可能的实现方式中，所述脑波数据用于反映所述用户获取到所述环境音频信号的神经元状态，所述目标音频信号是指所述环境音频信号中所述用户的神经元状态满足预设条件的音频信号。
其中，预设条件可以包括用户的神经元状态的波动范围满足预设阈值，或者，其它预先配置的条件；比如，使得神经元状态波动较大的音频信号可以为目标音频信号。
示例性地，可以将用户的脑波数据转换为声道运动信息，声道运动信息用于表示用户讲话时声道咬合部位的运动信息；根据声道运动信息得到目标音频信号。
在本申请的实施例中，可以根据用户的脑波数据获取用户关注的目标音频信号。具体流程可以参见后续图8所示的基于用户脑波数据获取目标音频信息的方法的示意图。
可选地，在一种可能的实现方式中，所述目标音频信号是指根据所述用户的行为日志预先配置的音频信号。
示例性地，用户的行为日志包括用户的兴趣爱好、用户的下载历史信息等。
可选地，在一种可能的实现方式中，目标音频信号为当前时刻的目标音频信号，还包括：
根据当前时刻的目标音频信号预测当前时刻的下一时刻的目标音频信号。
在本申请的实施例中，由于通过脑波数据获取用户的目标音频信号可能无法满足部分实时性业务需求；因此，还可以采用预测模型获取用户的目标音频信息；预测模型用于基于历史时刻或者当前时刻用户的脑波数据以及用户的目标音频信息预测未来时刻用户的目标音频信息。
应理解，当前时刻与当前时刻的下一时刻可以是连续的时刻，或者，也可以是指不连续的时刻；比如，当前时刻的下一时刻可以与当前时刻相差周期性时间间隔，该时间间隔的时间单位可以包括毫秒或者微秒。
可选地，在一种可能的实现方式中，获取环境音频信号，包括：
接收来自协作设备的环境音频信号，协作设备用于从声源处获取环境音频信号。
在本申请的实施例中，通过采用协作设备对采集的环境音频信号进行转发可以使得在环境音频信号在到达用户之前，用户就可以感知到这些信号；从而能够更高效的获取用户的关注的目标音频信息。具体流程可以参见后续图7所示。
S220、根据目标音频信号与初始参考信号，得到目标参考信号。
其中，目标参考信号中不包括目标音频信号。
可选地，在一种可能的实现方式中，根据目标音频信号与初始参考信号，得到目标参考信号，包括：
根据目标音频信号对初始参考信号进行滤波处理，得到目标参考信号。
需要说明的是，初始参考信号用于表示对环境音频信号进行主动降噪处理的噪声信号，即在初始参考信号中包括用户关注的音频信号与用户不关注的音频信号；通过对初始参考信号进行滤波处理，得到的目标参考信号中可以将用户不关注的音频信号进行滤除。
可选地，在一种可能的实现方式中，根据目标音频信号对初始参考信号进行滤波处理，包括：
根据目标音频信号通过采用滤波器对初始参考信号进行滤波处理。其中，滤波器可以包括自适应滤波器，或者，其它滤波器；具体流程可以参见后续图9所示。
S230、根据目标参考信号与环境音频信号，得到降噪信号。
其中，降噪信号用于抵消目标参考信号。比如，降噪信号可以产生反向波从而抵消目标参考信号。
进一步，在步骤S230之后，可以向用户播放降噪信号。
示例性地，图6是本申请实施例提供的主动降噪方法的架构图。图6所示的架构图中包括以下步骤：
步骤一：获取用户的目标音频信息。
其中，用户的目标音频信号可以是指用户感兴趣的音频信号，或者，可以是指与用户需求相关的音频信号。
在一个示例中，可以基于用户的脑波数据获取用户的目标音频信号。
例如，可以基于用户当前的脑波数据获取用户当前正在关注或者感兴趣的音频信号。
进一步地，为了更好的从用户的脑波数据中获取用户的目标音频信息，可以将外部的环境音频信号更快速地发送给用户；比如，通过采用协作设备将采集到的环境音频信号发送至降噪装置。
示例性地，如图7所示，协作设备可以是指独立于降噪装置的设备，比如，协作设备可以是指具有音频信息传递功能的电子设备；协作设备比可以包括手机、手表等各类电子产品。
例如，协作设备可以是一个设备或者多个设备；协作设备可以向降噪装置上报位置信息；上报位置信息的时机可以包括：在向降噪装置发送环境音频信号时上报位置信息，或者，也可以指协作设备在与降噪装置首次建立连接时上报位置信息，或者，可以将协作设备设置在默认位置/要求位置。
在本申请的实施例中，通过采用协作设备对采集的环境音频信号进行转发可以使得在环境音频信号在到达用户之前，用户就可以感知到这些信号；从而能够更高效的获取用户的关注的目标音频信息。
例如，可以基于用户当前的脑波数据获取用户当前正在关注或者感兴趣的音频信号。
示例性地，协作设备可以执行以下子步骤：
子步骤一：协作设备采集环境音频信号。
子步骤二：协作设备发送环境音频信号。
例如，协作设备可以将采集到的环境音频信号d(n)发送给降噪装置；比如，可以通过采用无线电的方式，如蓝牙、wifi、5G等方式将环境音频信号进行发送。可选地，在一种可能的实现方式中，协作设备可以将环境音频信号发送至降噪装置，在降噪装置上在播放环境音频信号之前，可以通过降噪装置对环境音频信号的能级进行缩放处理，使得播放声音不影响用户的正常感知；用于触发用户对关注的音频信息的提前感知，从而在直达声波d(n)到来之前形成这种音频感知，并基于脑波数据进行目标音频信息的呈现。
可选地，在另一种可能的实现方式中，也可以由协作设备对采集的环境音频信号进行缩放处理后再转发至降噪装置。
步骤二：采集环境音频信号。
例如，可以通过降噪装置中的麦克风设备采集用户所处环境的噪声信号x(n)，噪声信号x(n)中包括用户感兴趣的音频信号s(n)，以及相对于音频信号s(n)用户不关注的音频信号n(n)。
步骤三：音频信号滤波处理。
根据获取的用户的目标音频信号，从噪声信号x(n)中滤除目标音频信号，得到目标参考信号x’(n)。
需要说明的是，由于噪声信号x(n)中包括用户感兴趣的音频信号s(n)，以及相对于音频信号s(n)用户不关注的音频信号n(n)；对噪声信号x(n)进行滤波处理后，得到的目标参考信号x’(n)中仅包括用户不关注的音频信号n(n)部分。
步骤四：进行主动降噪处理。
根据目标参考信号x’(n)对采集的环境音频信号d(n)进行主动降噪处理，得到降噪信号。
应理解，环境音频信号d(n)为采集的直达降噪装置的音频信号，即d(n)中既包括用户关注的音频信号，也包括用户不关注的音频信号；根据目标参考信号对环境音频信号进行主动降噪处理，从而使得降噪处理后的降噪信号中包括用户关注的目标音频信息，实现用户更好的选择性聆听。
在一个示例中，上述步骤一中获取用户的目标音频信息的方法可以参见图8。
图8为基于用户脑波数据获取目标音频信号的方法的示意图。方法300包括步骤S310至步骤S330，下面分别对步骤S310至步骤S330进行详细的描述。
应理解，方法300的执行主体可以是降噪装置，或者，也可以是指独立于降噪装置的脑波数据处理装置，脑波数据处理装置用于获取用户的脑波数据进行分析并得到用户关注的目标音频信息。
S310、获取用户的脑波数据。
示例性地，用户的脑波数据可以通过脑波采集仪或其他脑波采集传感器进行采集得到的。
例如，用户的脑波数据可以是用户输入账户密码以登录医疗系统时进行采集，其包括不同时间点下的脑波信号值，或者，也可以是通过脑波采集仪实时获取的数据。当然，针对不同的应用场景，脑波数据也可以为在其他动作下进行采集，本实施例对此不做限定。
S320、将脑波数据转化为声道运动信息。
示例性地，为了实现脑波数据到声道运动信息的转化和映射，可以将人说话时的大量声道运动信息与用户的脑波数据(例如，神经活动数据相关联)；其中，声道运行信息可以包括嘴唇、舌头、喉和下颌的运动信息。
S330、根据声道运动信息得到目标音频信号。
示例性地，可以通过训练数据得到预先训练的循环神经网络得到声道运动信息对应的音频信息；其中，训练数据可以包括样本声道运动信息与样本音频信息，循环神经网络用于建立声道运行信息与音频信息之间的关联关系。
需要说明的是，图8所示的获取用户的目标音频信息的方法为举例说明，还可以采用现有技术中的其它方式通过脑波数据获取用户的目标音频信息，或者，也可以采用向降噪装置中预先配置用户的目标音频信息的方式，比如，目标音频信号可以是指预先配置的用户关注的文本信息，或者语音信息；本申请实施例对此不做任何限定。
在一个示例中，上述步骤三中音频信号滤波处理的方法可以参见图9。
例如，可以采用自适应滤波架构进行滤除处理，将获取的用户的目标音频信号从直接采集获取的噪声信号x(n)中去除，得到目标参考信号x’(n)。
应理解，图9所示的架构与ANC降噪算法模型原理类似。
例如，自适应滤波器的误差信号e(n)为：
e(n)＝x(n)-x'(n)＝x(n)-s' T(n)w(n)＝x(n)-w T(n)s'(n)；
其中，x(n)表示待滤波信号，即采集的初始参考信号；x'(n)表示滤波后输出的信号，即目标参考信号；s'(n)表示用户的目标音频信号，即用户关注的音频信号。
通过采用滤波算法后，可以将噪声信号x(n)中用户关注的目标音频信号s'(n)进行滤除，从而得到不包括用户关注的目标音频信号的目标参考信号x'(n)。
示例性地，上述图8对基于脑波数据获取目标音频信息的方法可以是通过实时采集用户的脑波数据进行分析获取用户的目标音频信息。
在一个示例中，由于通过脑波数据获取用户的目标音频信息可能无法满足部分实时性业务需求；因此，在本申请的实施例中，还可以采用预测模型获取用户的目标音频信息；预测模型用于基于历史时刻或者当前时刻用户的脑波数据以及用户的目标音频信息预测 未来时刻用户的目标音频信息。
例如，预测模型可以采用循环神经网络，循环神经网络的输出不仅依赖于当前的输入数据，同时也依赖于前一时刻的状态；通过时序反馈机制的引入有效地利用上下文信息，在语音、文本等时序信号的分析上有重要的意义。一个循环神经网络的输入可以由两部分组成，即前一时刻循环神经网络的隐含层状态信息和当前的输入信息，计算公式如下：
h t＝f(Wx t+Uh t-1+b)；
其中，W、U、b为模型的权重参数，
Figure PCTCN2020106681-appb-000002
式中h t表示t时刻的隐含层状态；x t表示t时刻的输入；f表示非线性激活函数，比如，可以采用tanh函数、sigmoid函数、ReLu函数等作为激活函数。
示例性地，循环神经网络可用于生成文本任务，假设输入X＝[x 1,x 2,x 3,x 4]，其对应的输出为Y＝[y 1,y 2,y 3,y 4]，对应公式如下：
Figure PCTCN2020106681-appb-000003
其中，V、c均为参数，
Figure PCTCN2020106681-appb-000004
表示t时刻循环神经网络的输出。
例如，在文本生成任务中，循环神经网络的输出维度大小可以与词表中单词个数相同。
在训练循环神经网络时，可以通过交叉熵作为损失函数进行训练，得到所需的语言模型，则预测模型可以基于历史时刻的用户的目标音频信息预测下一个时刻的用户的目标音频信息。在执行预测任务时，可以获取历史每个时刻的输出
Figure PCTCN2020106681-appb-000005
获取其中概率最大的词作为输出。
在本申请的实施例中可以使用的是语音信息，可以先将语音信息转换成文本信息；具体地，包括：
步骤一：将当前时刻获取的用户的目标音频信号s′(n)基于语音识别转换成语音文本text′(n)；
步骤二：将text′(n)作为循环神经网络的输入，然后基于预先训练的循环神经网络得到输出text″(n)，text″(n)可以用于表示预测的当前时刻的下一时刻的用户的目标音频信息；
步骤三：将text″(n)经过语音合成，得到预测的当前时刻的下一时刻的目标音频信息s″(n)；
步骤四：将s″(n)融入s′(n)，得到基于补充后预测的用户的目标音频信息s′(n)＝s′(n)+s″(n)。
进一步地，可以根据基于补充后预测的用户的目标音频信息采用图9所示的方法对采集的噪声信号x(n)中去除，得到目标参考信号x’(n)。
在本申请的实施例中，在基于脑波数据获取用户的目标音频信息时，即获取用户关注的音频信息时引入预测模型；从而可以进一步降低获取用户的目标音频信息的时间，根据用户的脑波数据与预测的用户的目标音频信息可以更好地了解用户关注的音频信息，使得用户关注的音频信息可以更好的保留在降噪信号中，从而实现用户更好地选择聆听。
在本申请中，通过获取用户关注的目标音频信号，从而得到目标参考信号；其中，目标参考信号中不包括用户的目标音频信号；根据目标参考信号与环境音频信号，得到降噪信号，降噪信号用于抵消目标参考信号；使得主动降噪处理后的降噪信号中依然包括用户关注的目标音频信号，从而使得主动降噪方法能够满足用户的个性化需求。
应理解，上述举例说明是为了帮助本领域技术人员理解本申请实施例，而非要将本申请实施例限于所例示的具体数值或具体场景。本领域技术人员根据所给出的上述举例说明，显然可以进行各种等价的修改或变化，这样的修改或变化也落入本申请实施例的范围内。
上文结合图1至图9，详细描述了本申请实施例提供的主动降噪方法；下面将结合图10至图12，详细描述本申请的装置实施例。应理解，本申请实施例中的主动降噪装置可以执行前述本申请实施例的各种主动降噪方法，即以下各种产品的具体工作过程，可以参考前述方法实施例中的对应过程。
图10是本申请提供的主动降噪装置的示意性框图。
应理解，主动降噪装置400可以执行图5至图9所示的主动降噪方法。该主动降噪装置400包括：获取模块410和处理模块420。
其中，获取模块410用于获取用户的目标音频信号、环境音频信号以及初始参考信号，其中，所述目标音频信号用于表示所述用户关注的音频信息，所述环境音频信号用于表示所述用户所处环境的音频信息，所述初始参考信号用于对所述环境音频信号进行主动降噪处理；处理模块420用于根据所述目标音频信号与所述初始参考信号，得到目标参考信号，所述目标参考信号中不包括所述目标音频信号；所述根据所述目标参考信号与所述环境音频信号，得到降噪信号，所述降噪信号用于抵消所述目标参考信号。
可选地，作为一个实施例，所述获取模块410具体用于：
获取所述用户的脑波数据；
所述处理模块具体用于：
根据所述用户的脑波数据获取所述用户的目标音频信号。
可选地，作为一个实施例，所述脑波数据用于反映所述用户获取到所述环境音频信号的神经元状态，所述目标音频信号是指所述环境音频信号中所述用户的神经元状态满足预设条件的音频信号。
可选地，作为一个实施例，所述目标音频信号是指根据所述用户的行为日志预先配置的音频信号。
可选地，作为一个实施例，所述处理模块420具体用于：
将所述用户的脑波数据转换为声道运动信息，所述声道运动信息用于表示所述用户讲话时声道咬合部位的运动信息；
根据所述声道运动信息得到所述目标音频信号。
可选地，作为一个实施例，所述目标音频信号为当前时刻的目标音频信号，所述处理模块420还用于：
根据所述当前时刻的目标音频信号预测所述当前时刻的下一时刻的目标音频信号。
可选地，作为一个实施例，所述处理模块420具体用于：
根据所述目标音频信号对所述初始参考信号进行滤波处理，得到所述目标参考信号。
可选地，作为一个实施例，所述处理模块420具体用于：
根据所述目标音频信号通过采用自适应滤波器对所述初始参考信号进行滤波处理。
可选地，作为一个实施例，所述获取模块410具体用于：
接收来自协作设备的所述环境音频信号，所述协作设备用于从声源处获取所述环境音 频信号。
在一个示例中，主动降噪装置400可以用于执行图5至图9中任意一个所示的方法中全部或部分操作。例如，获取模块可以用于执行S210、S310中全部或部分操作；处理模块可以用于执行S220、S230、S320、S330中全部或部分操作。其中，获取模块410可以是指主动降噪装置中的可以是通信接口或者收发器；比如，获取模块410可以是指主动降噪装置中的麦克风，或者接口电路。处理模块420可以是任意主动降噪装置中具有计算能力的处理器或芯片。
需要说明的是，上述主动降噪装置400以功能单元的形式体现。这里的术语“模块”可以通过软件和/或硬件形式实现，对此不作具体限定。
例如，“模块”可以是实现上述功能的软件程序、硬件电路或二者结合。所述硬件电路可能包括应用特有集成电路(application specific integrated circuit，ASIC)、电子电路、用于执行一个或多个软件或固件程序的处理器(例如共享处理器、专有处理器或组处理器等)和存储器、合并逻辑电路和/或其它支持所描述的功能的合适组件。
因此，在本申请的实施例中描述的各示例的单元，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
示例性，图11所示为主动降噪装置的示意图。其中，图10所示的获取模块410可以包括图11所示的脑波数据获取模块510以及环境音频信号获取模块520；处理模块420可以包括滤波处理模块530与降噪处理模块540。
其中，脑波数据获取模块510用于获取用户的脑波数据；根据脑波数据得到用户的目标音频信号。例如，脑波数据获取模块510可以用于执行S310至S330。
环境音频信号获取模块520用于获取用户所在的环境音频信号；比如，可以通过主动降噪装置的麦克风采集环境音频信号。例如，环境音频信号获取模块520可以用于执行S210。
滤波处理模块530用于从初始参考信号中对用户的目标音频信号进行滤波处理，得到目标参考信号。例如，环境音频信号获取模块520可以用于执行S220。
降噪处理模块540用于根据目标参考信号对环境音频信号进行主动降噪处理，得到降噪信号；其中，降噪信号中包括用户的目标音频信号，即用户依然可以听到用户关注的音频信息。例如，降噪处理模块540可以用于执行S230。
图12是本申请实施例提供的主动降噪装置的硬件结构示意图。
图12所示的主动降噪装置600包括存储器610、处理器620、通信接口630以及总线640。其中，存储器610、处理器620、通信接口630通过总线640实现彼此之间的通信连接。
存储器610可以是只读存储器(read only memory，ROM)，静态存储设备，动态存储设备或者随机存取存储器(random access memory，RAM)。存储器610可以存储程序，当存储器610中存储的程序被处理器620执行时，处理器620用于执行本申请实施例的主动降噪方法的各个步骤；例如，执行图5至图9所示的各个步骤。
应理解，本申请实施例所示的主动降噪装置可以是主动降噪耳机，也可以是配置于主 动降噪耳机中的芯片；或者，本申请实施例所示的主动降噪装置可以是车载头靠装置，也可以是配置于车载头靠装置中的芯片。
其中，主动降噪装置可以为具有主动降噪功能的设备，例如，可以包括当前技术已知的任何设备；或者，主动降噪装置还可以是指具有主动降噪功能的芯片。主动降噪装置中可以包括存储器和处理器；存储器可以用于存储程序代码，处理器可以用于调用存储器存储的程序代码，以实现计算设备的相应功能。计算设备中包括的处理器和存储器可以通过芯片实现，此处不作具体的限定。
例如，存储器可以用于存储本申请实施例中提供的主动降噪方法的相关程序指令，处理器可以用于调用存储器中存储的主动降噪方法的相关程序指令。
处理器620可以采用通用的中央处理器(central processing unit，CPU)，微处理器，应用专用集成电路(application specific integrated circuit，ASIC)，或者一个或多个集成电路，用于执行相关程序以实现本申请方法实施例的主动降噪方法。
处理器620还可以是一种集成电路芯片，具有信号的处理能力。在实现过程中，本申请的主动降噪方法的各个步骤可以通过处理器920中的硬件的集成逻辑电路或者软件形式的指令完成。
上述处理器620还可以是通用处理器、数字信号处理器(digital signal processing，DSP)、专用集成电路(ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器610，处理器620读取存储器610中的信息，结合其硬件完成本申请实施中图10或图11所示的主动降噪装置中包括的模块所需执行的功能，或者，执行本申请方法实施例的图5至图9所示的主动降噪方法。
通信接口630使用例如但不限于收发器一类的收发装置，来实现主动降噪装置600与其他设备或通信网络之间的通信。
总线640可包括在主动降噪装置600各个部件(例如，存储器610、处理器620、通信接口630)之间传送信息的通路。
应注意，尽管上述主动降噪装置600仅仅示出了存储器、处理器、通信接口，但是在具体实现过程中，本领域的技术人员应当理解，主动降噪装置600还可以包括实现正常运行所必须的其他器件。同时，根据具体需要本领域的技术人员应当理解，上述主动降噪装置600还可包括实现其他附加功能的硬件器件。此外，本领域的技术人员应当理解，上述主动降噪装置600也可仅仅包括实现本申请实施例所必须的器件，而不必包括图12中所示的全部器件。
示例性地，本申请实施例还提供一种主动降噪耳机，可以执行上述方法实施例中的主动降噪方法。
示例性地，本申请实施例还提供一种车载头靠装置，可以执行上述方法实施例中的主 动降噪方法。
例如，车载头靠装置可以是指汽车座椅的主动降噪头枕，在头枕本体表面可以开设有通孔，通孔中可以安装有一个或者多个降噪扬声器，降噪扬声器可以用于输出降噪信号，降噪信号可以是根据本申请实施例中的主动降噪方法得到的降噪信号。
在一个示例中，一个或者多个降噪扬声器可以安装在头枕中靠近用户的耳部位置，从而便于用户接收降噪扬声器输出降噪信号。在一个示例中，头枕本体上还可以安装有一个或者多个麦克风，麦克风可以用于采集环境音频信号或者初始参考信号；其中，用于采集环境音频信号或者初始参考信号的麦克风可以是同一个麦克风，也可以是指不同的麦克风。
例如，对于汽车而言，环境音频信号可以是用户在使用主动降噪头枕并未开启降噪功能时感知到的驾驶舱中的音频信息；初始参考信号可以是指用于对汽车的环境音频信号进行主动降噪处理的噪声信号；初始参考信号可以包括汽车的引擎噪声信号、风噪信号以及路噪信号中的一种或者多种。
在一个示例中，用于采集环境音频信号或者初始参考信号的麦克风可以安装在座舱内的任意位置。在一种可能的实现方式中，汽车座椅的主动降噪头枕可以是指安装在汽车驾驶员的座椅的主动降噪头枕；或者，也可以是指安装在汽车的乘客座椅的主动降噪头枕；本申请实施例对主动降噪头枕的具体安装位置不作任何限定。
示例性地，本申请实施例还提供一种汽车，包括上述方法实施例中的主动降噪装置。
例如，汽车中可以包括上述实施例中的主动降噪头枕。
示例性地，本申请实施例还提供一种主动降噪系统，可以执行上述方法实施例中的主动降噪方法。
示例性地，本申请实施例还提供一种芯片，该芯片包括收发单元和处理单元。其中，收发单元可以是输入输出电路、通信接口；处理单元为该芯片上集成的处理器或者微处理器或者集成电路；该芯片可以执行上述方法实施例中的主动降噪方法。
示例性地，本申请实施例还提供一种计算机可读存储介质，其上存储有指令，该指令被执行时执行上述方法实施例中的主动降噪方法。
示例性地，本申请实施例还提供一种包含指令的计算机程序产品，该指令被执行时执行上述方法实施例中的主动降噪方法。
应理解，本申请实施例中的处理器可以为中央处理单元(central processing unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
还应理解，本申请实施例中的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)， 其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的随机存取存储器(random access memory，RAM)可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
上述实施例，可以全部或部分地通过软件、硬件、固件或其他任意组合来实现。当使用软件实现时，上述实施例可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令或计算机程序。在计算机上加载或执行所述计算机指令或计算机程序时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以为通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集合的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质。半导体介质可以是固态硬盘。
应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况，其中A,B可以是单数或者复数。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系，但也可能表示的是一种“和/或”的关系，具体可参考前后文进行理解。
本申请中，“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a,b,或c中的至少一项(个)，可以表示：a,b,c,a-b,a-c,b-c,或a-b-c，其中a,b,c可以是单个，也可以是多个。
应理解，在本申请的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的 划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (25)

  1. 一种主动降噪方法，其特征在于，包括：
    获取用户的目标音频信号、环境音频信号以及初始参考信号，其中，所述目标音频信号用于表示所述用户关注的音频信息，所述环境音频信号用于表示所述用户所处环境的音频信息，所述初始参考信号用于对所述环境音频信号进行主动降噪处理；
    根据所述目标音频信号与所述初始参考信号，得到目标参考信号，所述目标参考信号中不包括所述目标音频信号；
    所述根据所述目标参考信号与所述环境音频信号，得到降噪信号，所述降噪信号用于抵消所述目标参考信号。
  2. 如权利要求1所述的方法，其特征在于，所述获取用户的目标音频信号，包括：
    获取所述用户的脑波数据；
    根据所述用户的脑波数据获取所述用户的目标音频信号。
  3. 如权利要求2所述的方法，其特征在于，所述脑波数据用于反映所述用户获取到所述环境音频信号的神经元状态，所述目标音频信号是指所述环境音频信号中所述用户的神经元状态满足预设条件的音频信号。
  4. 如权利要求1所述的方法，其特征在于，所述目标音频信号是指根据所述用户的行为日志预先配置的音频信号。
  5. 如权利要求2或3所述的方法，其特征在于，所述根据所述用户的脑波数据获取所述用户的目标音频信号，包括：
    将所述用户的脑波数据转换为声道运动信息，所述声道运动信息用于表示所述用户讲话时声道咬合部位的运动信息；
    根据所述声道运动信息得到所述目标音频信号。
  6. 如权利要求2至5中任一项所述的方法，其特征在于，所述目标音频信号为当前时刻的目标音频信号，还包括：
    根据所述当前时刻的目标音频信号预测所述当前时刻的下一时刻的目标音频信号。
  7. 如权利要求1至6中任一项所述的方法，其特征在于，所述根据所述目标音频信号与初始参考信号，得到目标参考信号，包括：
    根据所述目标音频信号对所述初始参考信号进行滤波处理，得到所述目标参考信号。
  8. 如权利要求7所述的方法，其特征在于，所述根据所述目标音频信号对所述初始参考信号进行滤波处理，包括：
    根据所述目标音频信号通过采用滤波器对所述初始参考信号进行滤波处理。
  9. 如权利要求1至8中任一项所述的方法，其特征在于，所述获取环境音频信号，包括：
    接收来自协作设备的所述环境音频信号，所述协作设备用于从声源处获取所述环境音频信号。
  10. 一种主动降噪装置，其特征在于，包括：
    获取模块，用于获取用户的目标音频信号、环境音频信号以及初始参考信号，其中， 所述目标音频信号用于表示所述用户关注的音频信息，所述环境音频信号用于表示所述用户所处环境的音频信息，所述初始参考信号用于对所述环境音频信号进行主动降噪处理；
    处理模块，用于根据所述目标音频信号与所述初始参考信号，得到目标参考信号，所述目标参考信号中不包括所述目标音频信号；所述根据所述目标参考信号与所述环境音频信号，得到降噪信号，所述降噪信号用于抵消所述目标参考信号。
  11. 如权利要求10所述的装置，其特征在于，所述获取模块具体用于：
    获取所述用户的脑波数据；
    所述处理模块具体用于：
    根据所述用户的脑波数据获取所述用户的目标音频信号。
  12. 如权利要求11所述的装置，其特征在于，所述脑波数据用于反映所述用户获取到所述环境音频信号的神经元状态，所述目标音频信号是指所述环境音频信号中所述用户的神经元状态满足预设条件的音频信号。
  13. 如权利要求10所述的装置，其特征在于，所述目标音频信号是指根据所述用户的行为日志预先配置的音频信号。
  14. 如权利要求11或12所述的装置，其特征在于，所述处理模块具体用于：
    将所述用户的脑波数据转换为声道运动信息，所述声道运动信息用于表示所述用户讲话时声道咬合部位的运动信息；
    根据所述声道运动信息得到所述目标音频信号。
  15. 如权利要求10至14中任一项所述的装置，其特征在于，所述目标音频信号为当前时刻的目标音频信号，所述处理模块还用于：
    根据所述当前时刻的目标音频信号预测所述当前时刻的下一时刻的目标音频信号。
  16. 如权利要求10至15中任一项所述的装置，其特征在于，所述处理模块具体用于：
    根据所述目标音频信号对所述初始参考信号进行滤波处理，得到所述目标参考信号。
  17. 如权利要求16所述的装置，其特征在于，所述处理模块具体用于：
    根据所述目标音频信号通过采用滤波器对所述初始参考信号进行滤波处理。
  18. 如权利要求10至17中任一项所述的装置，其特征在于，所述获取模块具体用于：
    接收来自协作设备所述环境音频信号，所述协作设备用于从声源处获取所述环境音频信号。
  19. 一种主动降噪耳机，其特征在于，用于执行根据权利要求1至9中任一项所述的方法。
  20. 一种车载头靠装置，其特征在于，用于执行根据权利要求1至9中任一项所述的方法。
  21. 一种汽车，其特征在于，包括如权利要求10至18中任一项所述的主动降噪装置。
  22. 一种主动降噪系统，其特征在于，包括如权利要求10至18中任一项所述的主动降噪装置。
  23. 一种主动降噪装置，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于执行所述存储器中存储的计算机程序，以使得所述主动降噪装置执行根据权利要求1至9中任一项所述的方法。
  24. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有程序指令，当所述程序指令由处理器运行时，实现权利要求1至9中任一项所述的方法。
  25. 一种芯片，其特征在于，所述芯片包括处理器与数据接口，所述处理器通过所述数据接口读取存储器上存储的指令，以执行如权利要求1至9中任一项所述的方法。
PCT/CN2020/106681 2020-08-04 2020-08-04 主动降噪方法、主动降噪装置以及主动降噪系统 WO2022027208A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2020/106681 WO2022027208A1 (zh) 2020-08-04 2020-08-04 主动降噪方法、主动降噪装置以及主动降噪系统
CN202080005894.7A CN114391166A (zh) 2020-08-04 2020-08-04 主动降噪方法、主动降噪装置以及主动降噪系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/106681 WO2022027208A1 (zh) 2020-08-04 2020-08-04 主动降噪方法、主动降噪装置以及主动降噪系统

Publications (1)

Publication Number Publication Date
WO2022027208A1 true WO2022027208A1 (zh) 2022-02-10

Family

ID=80119808

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/106681 WO2022027208A1 (zh) 2020-08-04 2020-08-04 主动降噪方法、主动降噪装置以及主动降噪系统

Country Status (2)

Country Link
CN (1) CN114391166A (zh)
WO (1) WO2022027208A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024016609A1 (zh) * 2022-07-20 2024-01-25 科大讯飞(苏州)科技有限公司 一种主动降噪方法、系统及相关装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115116458B (zh) * 2022-06-10 2024-03-08 腾讯科技（深圳）有限公司 语音数据转换方法、装置、计算机设备及存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104616662A (zh) * 2015-01-27 2015-05-13 中国科学院理化技术研究所 主动降噪方法及装置
CN107533839A (zh) * 2015-12-17 2018-01-02 华为技术有限公司 一种对周围环境音的处理方法及设备
CN107864440A (zh) * 2016-07-08 2018-03-30 奥迪康有限公司 包括eeg记录和分析系统的助听系统
CN108847250A (zh) * 2018-07-11 2018-11-20 会听声学科技（北京）有限公司 一种定向降噪方法、系统及耳机
CN109346053A (zh) * 2018-09-27 2019-02-15 珠海格力电器股份有限公司 降噪装置、控制方法及控制装置
WO2019111050A2 (en) * 2017-12-07 2019-06-13 Hed Technologies Sarl Voice aware audio system and method
CN110234050A (zh) * 2018-03-05 2019-09-13 哈曼国际工业有限公司 基于关注水平控制察觉的环境声音
CN110610719A (zh) * 2018-06-14 2019-12-24 奥迪康有限公司 声音处理设备

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104616662A (zh) * 2015-01-27 2015-05-13 中国科学院理化技术研究所 主动降噪方法及装置
CN107533839A (zh) * 2015-12-17 2018-01-02 华为技术有限公司 一种对周围环境音的处理方法及设备
CN107864440A (zh) * 2016-07-08 2018-03-30 奥迪康有限公司 包括eeg记录和分析系统的助听系统
WO2019111050A2 (en) * 2017-12-07 2019-06-13 Hed Technologies Sarl Voice aware audio system and method
CN110234050A (zh) * 2018-03-05 2019-09-13 哈曼国际工业有限公司 基于关注水平控制察觉的环境声音
CN110610719A (zh) * 2018-06-14 2019-12-24 奥迪康有限公司 声音处理设备
CN108847250A (zh) * 2018-07-11 2018-11-20 会听声学科技（北京）有限公司 一种定向降噪方法、系统及耳机
CN109346053A (zh) * 2018-09-27 2019-02-15 珠海格力电器股份有限公司 降噪装置、控制方法及控制装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024016609A1 (zh) * 2022-07-20 2024-01-25 科大讯飞(苏州)科技有限公司 一种主动降噪方法、系统及相关装置

Also Published As

Publication number Publication date
CN114391166A (zh) 2022-04-22

Similar Documents

Publication Publication Date Title
US11749262B2 (en) Keyword detection method and related apparatus
US11501772B2 (en) Context aware hearing optimization engine
US11910163B2 (en) Signal processing device, system and method for processing audio signals
WO2017101067A1 (zh) 一种对周围环境音的处理方法及设备
CN114556972A (zh) 用于辅助选择性听觉的系统和方法
WO2022027208A1 (zh) 主动降噪方法、主动降噪装置以及主动降噪系统
US11877125B2 (en) Method, apparatus and system for neural network enabled hearing aid
US11818547B2 (en) Method, apparatus and system for neural network hearing aid
US20220093118A1 (en) Signal processing device, system and method for processing audio signals
US11832061B2 (en) Method, apparatus and system for neural network hearing aid
US20230232170A1 (en) Method, Apparatus and System for Neural Network Hearing Aid
WO2022135340A1 (zh) 一种主动降噪的方法、设备及系统
CN109389990A (zh) 加强语音的方法、系统、车辆和介质
CN112767908B (zh) 基于关键声音识别的主动降噪方法、电子设备及存储介质
WO2022066393A1 (en) Hearing augmentation and wearable system with localized feedback
CN111654806B (zh) 音频播放方法、装置、存储介质及电子设备
CN115482830A (zh) 语音增强方法及相关设备
US20230209283A1 (en) Method for audio signal processing on a hearing system, hearing system and neural network for audio signal processing
WO2023136835A1 (en) Method, apparatus and system for neural network hearing aid
US11508388B1 (en) Microphone array based deep learning for time-domain speech signal extraction
CN117896469B (zh) 音频分享方法、装置、计算机设备和存储介质
Piazza et al. Digital Signal Processing for Audio Applications: Then, Now and the Future
CN117896469A (zh) 音频分享方法、装置、计算机设备和存储介质
Every et al. A Software-Centric Solution to Automotive Audio for General Purpose CPUs

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20948712

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20948712

Country of ref document: EP

Kind code of ref document: A1