Bluetooth earphone noise reduction method and system

Abstract

The invention relates to the technical field of Bluetooth headsets, and discloses a Bluetooth headset noise reduction method and system, and the method comprises the steps: carrying out the positioning of a Bluetooth headset, obtaining the position of the Bluetooth headset, and judging whether the position of the Bluetooth headset is changed or not; when the position of the earphone is not changed, analyzing planned environmental noise of the current environmental noise, performing noise phase inversion on the current environmental noise to obtain current phase inversion noise, and analyzing planned phase inversion noise of the current phase inversion noise; performing current noise subtraction on the current environment noise and the current anti-phase noise to obtain a current subtraction result, extracting a current subtraction parameter corresponding to the current subtraction result, and performing parameter adjustment on the current subtraction parameter by using the current subtraction result to obtain an adjusted current parameter; and acquiring playing voice of the Bluetooth headset at the headset position, and performing voice noise reduction on the playing voice based on the adjustment of the current parameter and the adjustment of the planning parameter to obtain a voice noise reduction result of the Bluetooth headset. According to the invention, the noise reduction complexity of the Bluetooth earphone can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of Bluetooth headsets, and in particular to a Bluetooth headset noise reduction method and system. Background Art

[0002] A Bluetooth headset is a device that transmits digital audio signals through wireless technology, converts sound signals into digital signals for transmission, and then restores them to analog signals through the digital-to-analog converter inside the headset. Bluetooth headsets can establish connections with other mobile devices and terminal devices, thereby receiving voice data sent by other mobile devices and terminal devices.

[0003] At present, the general principle of noise reduction of Bluetooth headsets is to collect noise data in the surrounding environment, invert the noise first, and cancel the inverted noise data with the voice data in the Bluetooth headset, thereby achieving noise reduction of the voice data in the Bluetooth headset. In this process, due to factors such as environmental uncertainty, when the inverted noise data and the voice data in the Bluetooth headset are canceled each other, it may appear that the mutual cancellation is incomplete, and multiple noise cancellation processes are required. The prior art has three problems in this regard. First, only a one-time noise cancellation process is performed. Second, multiple cancellations are performed but the same noise reduction parameters are used. Third, a complex optimization algorithm is used to calculate the noise reduction parameters for each round. In addition, when the Bluetooth headset is in the same environment for a long time, since the environmental noise in the same environment is relatively stable, there is no need to frequently collect the environmental noise of the Bluetooth headset. Therefore, the complexity of noise reduction of the Bluetooth headset is relatively high. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a Bluetooth headset noise reduction method and system, which can reduce the complexity of Bluetooth headset noise reduction.

[0005] In a first aspect, the present invention provides a Bluetooth headset noise reduction method, comprising:

[0006] Performing headset positioning on the Bluetooth headset to obtain the headset position, and determining whether the headset position has changed;

[0007] When the position of the earphone does not change, the current environmental noise of the Bluetooth earphone is collected, the planned environmental noise of the current environmental noise is analyzed, the current environmental noise is inverted to obtain the current inverted noise, and the planned inverted noise of the current inverted noise is analyzed;

[0008] Subtracting the current environmental noise from the current anti-phase noise to obtain a current subtraction result, extracting a current subtraction parameter corresponding to the current subtraction result, and adjusting the current subtraction parameter using the current subtraction result to obtain an adjusted current parameter;

[0009] Subtracting the planned environmental noise from the planned anti-phase noise to obtain a planned subtraction result, extracting a planned subtraction parameter corresponding to the planned subtraction result, and adjusting the planned subtraction parameter using the planned subtraction result to obtain an adjusted planning parameter;

[0010] The voice played by the Bluetooth headset at the headset position is collected, the current inverted noise and the planned inverted noise are spliced ​​to obtain the spliced ​​inverted noise, and based on the adjusted current parameters and the adjusted planned parameters, the spliced ​​inverted noise is used to perform voice noise reduction on the played voice to obtain the voice noise reduction result of the Bluetooth headset.

[0011] In a possible implementation manner of the first aspect, performing headset positioning on the Bluetooth headset to obtain the headset position includes:

[0012] When a Bluetooth connection is established between the Bluetooth headset and a preset mobile device, a positioning command is transmitted from the Bluetooth headset to the mobile device;

[0013] In the mobile device, based on the positioning command, a preset GPS positioning technology is used to detect the device location of the mobile device;

[0014] Determine a relative position of the Bluetooth headset with respect to the mobile device by using a Bluetooth connection between the Bluetooth headset and the mobile device;

[0015] The headset position of the Bluetooth headset is calculated based on the device position and the relative position.

[0016] In a possible implementation manner of the first aspect, determining whether the position of the earphone changes includes:

[0017] Query the noise collection distance of the Bluetooth headset;

[0018] The earphone position is taken as the area center, and the noise collection distance is taken as the area radius;

[0019] Determine the spatial region where the region center and the region radius are located;

[0020] identifying a noise source in the region of space;

[0021] Collecting noise indicators of the noise source;

[0022] The noise index includes the number of noise sources, noise level, and noise source-headphone distance;

[0023] Calculating the noise score in the spatial area under the noise index using a preset index scoring algorithm;

[0024] grading the noise scores to obtain score levels;

[0025] When the earphone position moves and the score level changes, determining that the earphone position changes;

[0026] When the position of the earphone moves and the score level does not change, it is determined that the position of the earphone does not change.

[0027] In a possible implementation manner of the first aspect, the analyzing the planned environmental noise of the current environmental noise includes:

[0028] The current environmental noise is decomposed into IMF components using the following formula:

[0029]

[0030] Among them, c i (t) represents the IMF component, x(t) represents the current environmental noise, r n (t) represents the residual term, n represents the number of IMF components, i represents the sequence number of the IMF component, and t represents the time;

[0031] The Hilbert transform is performed on the IMF component using the following formula to obtain a Hilbert transform result:

[0032]

[0033] Among them, y i (t) represents the Hilbert transform result, c i (t ′ ) indicates that the independent variable is t ′ The IMF component of , t represents the time;

[0034] Based on the IMF component and the Hilbert transform result, the instantaneous amplitude of the current environmental noise is calculated using the following formula:

[0035]

[0036] Among them, a i (t) represents the instantaneous amplitude, c i (t) represents the IMF component, y i (t) represents the Hilbert transform result;

[0037] Based on the IMF component and the Hilbert transform result, the instantaneous frequency of the current environmental noise is calculated using the following formula:

[0038]

[0039] Among them, ω i(t) represents the instantaneous frequency, θ i (t) represents the instantaneous phase, c i (t) represents the IMF component, y i (t) represents the Hilbert transform result;

[0040] The instantaneous amplitude and the instantaneous frequency are input into a preset long short-term memory neural network, so as to output the planned environmental noise corresponding to the instantaneous amplitude and the instantaneous frequency through the long short-term memory neural network.

[0041] In a possible implementation of the first aspect, the step of inputting the instantaneous amplitude and the instantaneous frequency into a preset long short-term memory neural network, so as to output the planned environmental noise corresponding to the instantaneous amplitude and the instantaneous frequency through the long short-term memory neural network, includes:

[0042] constructing an amplitude sequence of the instantaneous amplitude;

[0043] constructing a frequency sequence of the instantaneous frequency;

[0044] Calculating the amplitude weighted vector of the amplitude sequence using the attention mechanism in the long short-term memory neural network;

[0045] Calculating a frequency weighted vector of the frequency sequence using an attention mechanism in the long short-term memory neural network;

[0046] Using a long short-term memory layer in the long short-term memory neural network to capture the amplitude dependency in the amplitude weighted vector;

[0047] Using a long short-term memory layer in the long short-term memory neural network to capture the frequency dependency in the frequency weighted vector;

[0048] The amplitude dependency and the frequency dependency are linearly mapped into the planning environment noise using a fully connected layer in the long short-term memory neural network.

[0049] In a possible implementation manner of the first aspect, inverting the current environmental noise to obtain current inverted noise includes:

[0050] A noise inverter for determining said current ambient noise;

[0051] identifying an input terminal and an output terminal of the noise inverter;

[0052] Inputting the current environmental noise to the input terminal;

[0053] In the noise inverter, after receiving the current environmental noise from the input end, the current inverted noise is output at the output end.

[0054] In a possible implementation manner of the first aspect, performing current noise subtraction on the current environmental noise and the current anti-phase noise to obtain a current subtraction result includes:

[0055] Constructing a noise summing circuit between the current environmental noise and the current anti-phase noise;

[0056] Inputting the current environmental noise and the current anti-phase noise into the noise summing circuit;

[0057] In the noise summing circuit, a current noise sum between the current environmental noise and the current anti-phase noise is calculated;

[0058] The current noise sum is used as the current subtraction result.

[0059] In a possible implementation manner of the first aspect, the step of adjusting the current subtraction parameter by using the current subtraction result to obtain the adjusted current parameter includes:

[0060] constructing an error controller for the current subtraction result;

[0061] Calculating a subtraction result error between the current subtraction result and a preset reference subtraction result;

[0062] Inputting the subtraction result error into the error controller, so that the error controller outputs the target current parameter corresponding to the subtraction result error;

[0063] Wherein, the error controller comprises:

[0064]

[0065] Among them, u(t) represents the current parameter of the target, e(t) represents the subtraction result error, K p Represents the proportionality coefficient, K i Indicates the integral coefficient, K d represents the differential coefficient, t represents the time;

[0066] The current subtraction parameter is adjusted to the target current parameter to obtain the adjusted current parameter.

[0067] In a possible implementation of the first aspect, based on the adjusting current parameter and the adjusting planned parameter, using the spliced ​​inverted noise to perform voice noise reduction on the played voice, to obtain a voice noise reduction result of the Bluetooth headset, includes:

[0068] Obtaining a noise separation moment between the current inverted phase noise and the planned inverted phase noise in the spliced ​​inverted phase noise;

[0069] Based on the adjustment current parameters and the adjustment planning parameters, the noise separation moment is used to perform noise-voice subtraction between the spliced ​​inverted noise and the played voice to obtain the voice noise reduction result of the Bluetooth headset.

[0070] In a second aspect, the present invention provides a Bluetooth headset noise reduction system, the system comprising:

[0071] The headset positioning module is used to locate the Bluetooth headset, obtain the headset position, and determine whether the headset position has changed;

[0072] A noise inversion module, used for collecting the current environmental noise of the Bluetooth headset when the position of the headset does not change, analyzing the planned environmental noise of the current environmental noise, performing noise inversion on the current environmental noise to obtain the current inverted noise, and analyzing the planned inverted noise of the current inverted noise;

[0073] A current parameter adjustment module, used to perform current noise subtraction on the current environmental noise and the current anti-phase noise to obtain a current subtraction result, extract a current subtraction parameter corresponding to the current subtraction result, and perform parameter adjustment on the current subtraction parameter using the current subtraction result to obtain an adjusted current parameter;

[0074] A planning parameter adjustment module, used to perform planning noise subtraction on the planned environmental noise and the planned anti-phase noise to obtain a planning subtraction result, extract a planning subtraction parameter corresponding to the planning subtraction result, and perform parameter adjustment on the planning subtraction parameter using the planning subtraction result to obtain an adjusted planning parameter;

[0075] A speech noise reduction module is used to collect the played speech of the Bluetooth headset at the headset position, splice the current inverted noise and the planned inverted noise to obtain the spliced ​​inverted noise, and based on the adjusted current parameters and the adjusted planned parameters, use the spliced ​​inverted noise to perform speech noise reduction on the played speech to obtain the speech noise reduction result of the Bluetooth headset.

[0076] Compared with the prior art, the technical principle and beneficial effects of this solution are:

[0077] The embodiment of the present invention locates the Bluetooth headset to detect whether the Bluetooth headset is in the same noise environment for a long time. Furthermore, the embodiment of the present invention analyzes the planned environmental noise of the current environmental noise to predict the noise data of the future time period based on the neural network model, thereby reducing the workload of frequently collecting noise data of the surrounding environment and reducing the complexity of noise reduction of the Bluetooth headset. Furthermore, the embodiment of the present invention adjusts the current subtraction parameters by using the current subtraction result to perform multiple offsets when the noise data is offset against each other, and adjusts the noise reduction parameters of each round by the parameters output by the error controller, and replaces the target optimization algorithm with high complexity by the error controller, thereby reducing the complexity of speech noise reduction. Therefore, a Bluetooth headset noise reduction method and system proposed in the embodiment of the present invention can reduce the complexity of Bluetooth headset noise reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0079] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0080] Figure 1 A schematic diagram of a flow chart of a Bluetooth headset noise reduction method provided by an embodiment of the present invention;

[0081] Figure 2 A schematic diagram of a noise summing circuit for a Bluetooth headset noise reduction method provided by an embodiment of the present invention;

[0082] Figure 3 A schematic diagram of a module of a Bluetooth headset noise reduction system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0083] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0084] The embodiment of the present invention provides a Bluetooth headset noise reduction method, and the execution subject of the Bluetooth headset noise reduction method includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present invention. In other words, the Bluetooth headset noise reduction method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0085] See also Figure 1 FIG. 1 is a flow chart of a Bluetooth headset noise reduction method provided by an embodiment of the present invention. Figure 1 The Bluetooth headset noise reduction methods described in include:

[0086] S1. Positioning the Bluetooth headset to obtain the headset position, and determining whether the headset position changes.

[0087] The embodiment of the present invention performs earphone positioning on the Bluetooth earphone to detect whether the Bluetooth earphone is in the same noise environment for a long time.

[0088] In one embodiment of the present invention, the method of locating the Bluetooth headset to obtain the headset position includes: when a Bluetooth connection is established between the Bluetooth headset and a preset mobile device, transmitting a positioning command from the Bluetooth headset to the mobile device; in the mobile device, based on the positioning command, detecting the device position of the mobile device using a preset GPS positioning technology; determining the relative position of the Bluetooth headset with respect to the mobile device using the Bluetooth connection between the Bluetooth headset and the mobile device; and calculating the headset position of the Bluetooth headset based on the device position and the relative position.

[0089] The positioning command is mainly used to obtain the positioning authority of the mobile device.

[0090] In one embodiment of the present invention, the determining whether the position of the earphone has changed includes: querying the noise collection distance of the Bluetooth earphone; taking the earphone position as the area center and the noise collection distance as the area radius; determining the spatial area where the area center and the area radius are located; identifying the noise source in the spatial area; collecting the noise index of the noise source; wherein the noise index includes the number of noise sources, the noise level, and the noise source-earphone distance; calculating the noise score in the spatial area under the noise index using a preset index scoring algorithm; grading the noise score to obtain a score level; when the earphone position moves and the score level changes, determining that the earphone position has changed; when the earphone position moves and the score level does not change, determining that the earphone position has not changed.

[0091] Among them, the noise collection distance refers to the distance at which the Bluetooth headset collects the surrounding environmental noise, the noise source refers to the source of the noise in the spatial area, such as the sound of machines, vehicles, body movements, etc., and the indicator scoring algorithm can be an objective weighting method.

[0092] Optionally, the process of grading the noise scores to obtain score levels is to arrange the noise scores from small to large and then classify a plurality of similar noise scores as noise scores of the same level.

[0093] S2. When the position of the earphone does not change, collect the current environmental noise of the Bluetooth earphone, analyze the planned environmental noise of the current environmental noise, invert the current environmental noise to obtain the current inverted noise, and analyze the planned inverted noise of the current inverted noise.

[0094] When the position of the earphone has not changed, it means that the Bluetooth earphone has been in the same noise environment for a long time. At this time, there is no need to frequently collect noise data of the surrounding environment. Instead, the noise data of the future time period can be predicted based on the neural network model.

[0095] Furthermore, the embodiment of the present invention analyzes the planned environmental noise of the current environmental noise to predict the noise data of the future time period based on the neural network model, thereby reducing the workload of frequently collecting noise data of the surrounding environment and reducing the complexity of noise reduction of Bluetooth headsets.

[0096] The current environmental noise refers to the environmental noise in the current period, and the planned environmental noise refers to the environmental noise in the future period.

[0097] In one embodiment of the present invention, the analyzing the planned environmental noise of the current environmental noise includes: decomposing the current environmental noise into IMF components using the following formula:

[0098]

[0099] Among them, c i (t) represents the IMF component, x(t) represents the current environmental noise, r n (t) represents the residual term, n represents the number of IMF components, i represents the sequence number of the IMF component, and t represents the time;

[0100] The Hilbert transform is performed on the IMF component using the following formula to obtain a Hilbert transform result:

[0101]

[0102] Among them, y i (t) represents the Hilbert transform result, c i (t ′ ) indicates that the independent variable is t ′ The IMF component of , t represents the time;

[0103] Based on the IMF component and the Hilbert transform result, the instantaneous amplitude of the current environmental noise is calculated using the following formula:

[0104]

[0105] Among them, a i (t) represents the instantaneous amplitude, c i (t) represents the IMF component, y i (t) represents the Hilbert transform result;

[0106] Based on the IMF component and the Hilbert transform result, the instantaneous frequency of the current environmental noise is calculated using the following formula:

[0107]

[0108] Among them, ω i (t) represents the instantaneous frequency, θ i (t) represents the instantaneous phase, c i (t) represents the IMF component, y i (t) represents the Hilbert transform result;

[0109] The instantaneous amplitude and the instantaneous frequency are input into a preset long short-term memory neural network, so as to output the planned environmental noise corresponding to the instantaneous amplitude and the instantaneous frequency through the long short-term memory neural network.

[0110] In another embodiment of the present invention, the instantaneous amplitude and the instantaneous frequency are input into a preset long short-term memory neural network to output the planned environmental noise corresponding to the instantaneous amplitude and the instantaneous frequency through the long short-term memory neural network, including: constructing an amplitude sequence of the instantaneous amplitude; constructing a frequency sequence of the instantaneous frequency; using the attention mechanism in the long short-term memory neural network to calculate the amplitude weighted vector of the amplitude sequence; using the attention mechanism in the long short-term memory neural network to calculate the frequency weighted vector of the frequency sequence; using the long short-term memory layer in the long short-term memory neural network to capture the amplitude dependency in the amplitude weighted vector; using the long short-term memory layer in the long short-term memory neural network to capture the frequency dependency in the frequency weighted vector; using the fully connected layer in the long short-term memory neural network to linearly map the amplitude dependency and the frequency dependency to the planned environmental noise.

[0111] Among them, the amplitude sequence and the frequency sequence are both sequences arranged in ascending order of time, the attention mechanism can be a bidirectional attention mechanism and a multi-head attention mechanism, etc., the amplitude weighted vector and the frequency weighted vector are both output results of the attention mechanism, and the amplitude dependency and the frequency dependency are both output results of the long short-term memory layer.

[0112] In one embodiment of the present invention, the current environmental noise is subjected to noise inversion to obtain the current inverted noise, comprising: determining a noise inverter for the current environmental noise; identifying an input end and an output end of the noise inverter; inputting the current environmental noise to the input end; and in the noise inverter, after receiving the current environmental noise from the input end, outputting the current inverted noise at the output end.

[0113] S3. Subtract the current environmental noise from the current anti-phase noise to obtain a current subtraction result, extract a current subtraction parameter corresponding to the current subtraction result, and use the current subtraction result to adjust the current subtraction parameter to obtain an adjusted current parameter.

[0114] In one embodiment of the present invention, performing current noise subtraction between the current environmental noise and the current inverted noise to obtain a current subtraction result includes: constructing a noise summing circuit between the current environmental noise and the current inverted noise; inputting the current environmental noise and the current inverted noise into the noise summing circuit; calculating the current noise sum between the current environmental noise and the current inverted noise in the noise summing circuit; and using the current noise sum as the current subtraction result.

[0115] See also Figure 2FIG. 1 is a schematic diagram of a noise summing circuit of a Bluetooth headset noise reduction method provided by an embodiment of the present invention. Figure 2 In the figure, 1 represents the current environmental noise, 2 represents the current anti-phase noise, R1 and R2 represent the current subtraction parameters, and 3 represents the current subtraction result.

[0116] Furthermore, the embodiment of the present invention adjusts the current subtraction parameters by utilizing the current subtraction result so as to perform multiple cancellations when the noise data are cancelled out from each other, and adjusts the noise reduction parameters of each round through the parameters output by the error controller. The error controller is used to replace the highly complex target optimization algorithm, thereby reducing the complexity of speech noise reduction.

[0117] In one embodiment of the present invention, the method of adjusting the current subtraction parameter using the current subtraction result to obtain the adjusted current parameter includes: constructing an error controller of the current subtraction result; calculating a subtraction result error between the current subtraction result and a preset reference subtraction result; inputting the subtraction result error into the error controller, so as to output a target current parameter corresponding to the subtraction result error through the error controller; wherein the error controller includes:

[0118]

[0119] Among them, u(t) represents the current parameter of the target, e(t) represents the subtraction result error, K p Represents the proportionality coefficient, K i Indicates the integral coefficient, K f represents the differential coefficient, t represents the time;

[0120] The current subtraction parameter is adjusted to the target current parameter to obtain the adjusted current parameter.

[0121] The value of the reference subtraction result is 0.

[0122] S4. Subtract the planned environmental noise from the planned anti-phase noise to obtain a planned subtraction result, extract a planned subtraction parameter corresponding to the planned subtraction result, and adjust the planned subtraction parameter using the planned subtraction result to obtain an adjusted planning parameter.

[0123] Optionally, the process of performing planned noise subtraction on the planned environment noise and the planned inverse noise to obtain the planned subtraction result is similar to the principle of performing current noise subtraction on the current environment noise and the current inverse noise to obtain the current subtraction result, and is not further elaborated here.

[0124] Optionally, the process of using the planned subtraction result to adjust the planned subtraction parameters to obtain the adjusted planned parameters is similar to the aforementioned principle of using the current subtraction result to adjust the current subtraction parameters to obtain the adjusted current parameters, and will not be further elaborated here.

[0125] S5. Collect the played voice of the Bluetooth headset at the headset position, splice the current inverted noise and the planned inverted noise to obtain the spliced ​​inverted noise, and perform voice noise reduction on the played voice using the spliced ​​inverted noise based on the adjusted current parameters and the adjusted planned parameters to obtain the voice noise reduction result of the Bluetooth headset.

[0126] In one embodiment of the present invention, based on the adjustment of the current parameters and the adjustment of the planned parameters, the spliced ​​inverted noise is used to perform voice noise reduction on the played voice to obtain the voice noise reduction result of the Bluetooth headset, including: obtaining the noise separation time between the current inverted noise and the planned inverted noise in the spliced ​​inverted noise; based on the adjustment of the current parameters and the adjustment of the planned parameters, using the noise separation time to perform noise-voice subtraction between the spliced ​​inverted noise and the played voice to obtain the voice noise reduction result of the Bluetooth headset.

[0127] Among them, the noise separation moment refers to the last moment of the sequence of the current inverse noise. According to the noise separation moment, it can be identified which speech sequence is the speech sequence corresponding to the current inverse noise and which speech sequence is the speech sequence corresponding to the planned inverse noise.

[0128] Optionally, the process of performing noise-voice subtraction between the spliced ​​inverted noise and the played voice based on the adjustment current parameters and the adjustment planned parameters and utilizing the noise separation moment to obtain the voice noise reduction result of the Bluetooth headset is similar to the above-mentioned principle of performing current noise subtraction between the current ambient noise and the current inverted noise to obtain the current subtraction result, and will not be further elaborated here.

[0129] It can be seen that the embodiment of the present invention locates the Bluetooth headset to detect whether the Bluetooth headset is in the same noise environment for a long time. Furthermore, the embodiment of the present invention analyzes the planned environmental noise of the current environmental noise to predict the noise data of the future time period based on the neural network model, thereby reducing the workload of frequently collecting noise data of the surrounding environment and reducing the complexity of Bluetooth headset noise reduction. Furthermore, the embodiment of the present invention uses the current subtraction result to adjust the current subtraction parameter to cancel out the noise data multiple times, and adjusts the noise reduction parameters of each round through the parameters output by the error controller, and replaces the target optimization algorithm with high complexity through the error controller, thereby reducing the complexity of speech noise reduction. Therefore, the Bluetooth headset noise reduction method and system proposed in the embodiment of the present invention can reduce the complexity of Bluetooth headset noise reduction.

[0130] like Figure 3 , which is a functional module diagram of the Bluetooth headset noise reduction system of the present invention.

[0131] The Bluetooth headset noise reduction system 300 of the present invention can be installed in an electronic device. According to the functions to be implemented, the Bluetooth headset noise reduction system can include a headset positioning module 301, a noise inversion module 302, a current parameter adjustment module 303, a planning parameter adjustment module 304 and a speech noise reduction module 305. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0132] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0133] The headset positioning module 301 is used to locate the Bluetooth headset, obtain the headset position, and determine whether the headset position has changed;

[0134] The noise inversion module 302 is used to collect the current environmental noise of the Bluetooth headset when the position of the headset does not change, analyze the planned environmental noise of the current environmental noise, invert the current environmental noise to obtain the current inverted noise, and analyze the planned inverted noise of the current inverted noise;

[0135] The current parameter adjustment module 303 is used to perform current noise subtraction on the current environmental noise and the current anti-phase noise to obtain a current subtraction result, extract a current subtraction parameter corresponding to the current subtraction result, and perform parameter adjustment on the current subtraction parameter using the current subtraction result to obtain an adjusted current parameter;

[0136] The planning parameter adjustment module 304 is used to perform planning noise subtraction on the planned environmental noise and the planned anti-phase noise to obtain a planning subtraction result, extract a planning subtraction parameter corresponding to the planning subtraction result, and perform parameter adjustment on the planning subtraction parameter using the planning subtraction result to obtain an adjusted planning parameter;

[0137] The speech noise reduction module 305 is used to collect the played speech of the Bluetooth headset at the headset position, splice the current inverted noise and the planned inverted noise to obtain the spliced ​​inverted noise, and based on the adjusted current parameters and the adjusted planned parameters, use the spliced ​​inverted noise to perform speech noise reduction on the played speech to obtain the speech noise reduction result of the Bluetooth headset.

[0138] In detail, each module in the Bluetooth headset noise reduction system 300 in the embodiment of the present invention is used in the same manner as described above. Figure 1 The Bluetooth headset noise reduction method described in the present invention has the same technical means and can produce the same technical effects, so I will not go into details here.

[0139] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0140] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0141] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0142] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0143] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A Bluetooth headset noise reduction method, characterized in that: The method comprises: Performing headset positioning on the Bluetooth headset to obtain the headset position, and determining whether the headset position has changed; When the position of the earphone does not change, the current environmental noise of the Bluetooth earphone is collected, the planned environmental noise of the current environmental noise is analyzed, the current environmental noise is inverted to obtain the current inverted noise, and the planned inverted noise of the current inverted noise is analyzed; Subtracting the current environmental noise from the current anti-phase noise to obtain a current subtraction result, extracting a current subtraction parameter corresponding to the current subtraction result, and adjusting the current subtraction parameter using the current subtraction result to obtain an adjusted current parameter; Subtracting the planned environmental noise from the planned anti-phase noise to obtain a planned subtraction result, extracting a planned subtraction parameter corresponding to the planned subtraction result, and adjusting the planned subtraction parameter using the planned subtraction result to obtain an adjusted planning parameter; The voice played by the Bluetooth headset at the headset position is collected, the current inverted noise and the planned inverted noise are spliced ​​to obtain the spliced ​​inverted noise, and based on the adjusted current parameters and the adjusted planned parameters, the spliced ​​inverted noise is used to perform voice noise reduction on the played voice to obtain the voice noise reduction result of the Bluetooth headset.

2. The method according to claim 1, characterized in that The method of locating the Bluetooth headset to obtain the headset position includes: When a Bluetooth connection is established between the Bluetooth headset and a preset mobile device, a positioning command is transmitted from the Bluetooth headset to the mobile device; In the mobile device, based on the positioning command, a preset GPS positioning technology is used to detect the device location of the mobile device; Determine a relative position of the Bluetooth headset with respect to the mobile device by using a Bluetooth connection between the Bluetooth headset and the mobile device; The headset position of the Bluetooth headset is calculated based on the device position and the relative position.

3. The method according to claim 1, characterized in that: The determining whether the position of the earphone has changed includes: Query the noise collection distance of the Bluetooth headset; The earphone position is taken as the area center, and the noise collection distance is taken as the area radius; Determine the spatial region where the region center and the region radius are located; identifying a noise source in the region of space; Collecting noise indicators of the noise source; The noise index includes the number of noise sources, noise level, and noise source-headphone distance; Calculating the noise score in the spatial area under the noise index using a preset index scoring algorithm; grading the noise scores to obtain score levels; When the earphone position moves and the score level changes, determining that the earphone position changes; When the position of the earphone moves and the score level does not change, it is determined that the position of the earphone does not change.

4. The method according to claim 1, characterized in that: The analyzing the planned environmental noise of the current environmental noise includes: The current environmental noise is decomposed into IMF components using the following formula: Among them, c i (t) represents the IMF component, x(t) represents the current environmental noise, r n (t) represents the residual term, n represents the number of IMF components, i represents the sequence number of the IMF component, and t represents the time; The Hilbert transform is performed on the IMF component using the following formula to obtain a Hilbert transform result: Among them, y i (t) represents the Hilbert transform result, c i (t ′ ) indicates that the independent variable is t ′ The IMF component of , t represents the time; Based on the IMF component and the Hilbert transform result, the instantaneous amplitude of the current environmental noise is calculated using the following formula: Among them, a i (t) represents the instantaneous amplitude, c i (t) represents the IMF component, y i (t) represents the Hilbert transform result; Based on the IMF component and the Hilbert transform result, the instantaneous frequency of the current environmental noise is calculated using the following formula: Among them, ω i (t) represents the instantaneous frequency, θ i (t) represents the instantaneous phase, c i (t) represents the IMF component, y i (t) represents the Hilbert transform result; The instantaneous amplitude and the instantaneous frequency are input into a preset long short-term memory neural network, so as to output the planned environmental noise corresponding to the instantaneous amplitude and the instantaneous frequency through the long short-term memory neural network.

5. The method according to claim 4, characterized in that The step of inputting the instantaneous amplitude and the instantaneous frequency into a preset long short-term memory neural network, so as to output the planned environmental noise corresponding to the instantaneous amplitude and the instantaneous frequency through the long short-term memory neural network, comprises: constructing an amplitude sequence of the instantaneous amplitude; constructing a frequency sequence of the instantaneous frequency; Calculating the amplitude weighted vector of the amplitude sequence using the attention mechanism in the long short-term memory neural network; Calculating a frequency weighted vector of the frequency sequence using an attention mechanism in the long short-term memory neural network; Using a long short-term memory layer in the long short-term memory neural network to capture the amplitude dependency in the amplitude weighted vector; Using a long short-term memory layer in the long short-term memory neural network to capture the frequency dependency in the frequency weighted vector; The amplitude dependency and the frequency dependency are linearly mapped into the planning environment noise using a fully connected layer in the long short-term memory neural network.

6. The method according to claim 1, characterized in that The step of inverting the current environmental noise to obtain current inverted noise includes: A noise inverter for determining said current ambient noise; identifying an input terminal and an output terminal of the noise inverter; Inputting the current environmental noise to the input terminal; In the noise inverter, after receiving the current environmental noise from the input end, the current inverted noise is output at the output end.

7. The method according to claim 1, characterized in that The subtracting the current environmental noise from the current anti-phase noise to obtain a current subtraction result includes: Constructing a noise summing circuit between the current environmental noise and the current anti-phase noise; Inputting the current environmental noise and the current anti-phase noise into the noise summing circuit; In the noise summing circuit, a current noise sum between the current environmental noise and the current anti-phase noise is calculated; The current noise sum is used as the current subtraction result.

8. The method according to claim 1, characterized in that The step of adjusting the current subtraction parameter by using the current subtraction result to obtain the adjusted current parameter includes: constructing an error controller for the current subtraction result; Calculating a subtraction result error between the current subtraction result and a preset reference subtraction result; Inputting the subtraction result error into the error controller, so that the error controller outputs the target current parameter corresponding to the subtraction result error; Wherein, the error controller comprises: Among them, u(t) represents the current parameter of the target, e(t) represents the subtraction result error, K p Represents the proportionality coefficient, K i Indicates the integral coefficient, K d represents the differential coefficient, t represents the time; The current subtraction parameter is adjusted to the target current parameter to obtain the adjusted current parameter.

9. The method according to claim 1, characterized in that: The method of performing voice noise reduction on the played voice by using the spliced ​​inverted noise based on the adjusted current parameter and the adjusted planned parameter to obtain the voice noise reduction result of the Bluetooth headset includes: Obtaining a noise separation moment between the current inverted phase noise and the planned inverted phase noise in the spliced ​​inverted phase noise; Based on the adjustment current parameters and the adjustment planning parameters, the noise separation moment is used to perform noise-voice subtraction between the spliced ​​inverted noise and the played voice to obtain the voice noise reduction result of the Bluetooth headset.

10. A Bluetooth headset noise reduction system, characterized in that: The system comprises: The headset positioning module is used to locate the Bluetooth headset, obtain the headset position, and determine whether the headset position has changed; A noise inversion module, used for collecting the current environmental noise of the Bluetooth headset when the position of the headset does not change, analyzing the planned environmental noise of the current environmental noise, performing noise inversion on the current environmental noise to obtain the current inverted noise, and analyzing the planned inverted noise of the current inverted noise; A current parameter adjustment module, used to perform current noise subtraction on the current environmental noise and the current anti-phase noise to obtain a current subtraction result, extract a current subtraction parameter corresponding to the current subtraction result, and perform parameter adjustment on the current subtraction parameter using the current subtraction result to obtain an adjusted current parameter; A planning parameter adjustment module, used to perform planning noise subtraction on the planned environmental noise and the planned anti-phase noise to obtain a planning subtraction result, extract a planning subtraction parameter corresponding to the planning subtraction result, and perform parameter adjustment on the planning subtraction parameter using the planning subtraction result to obtain an adjusted planning parameter; A speech noise reduction module is used to collect the played speech of the Bluetooth headset at the headset position, splice the current inverted noise and the planned inverted noise to obtain the spliced ​​inverted noise, and based on the adjusted current parameters and the adjusted planned parameters, use the spliced ​​inverted noise to perform speech noise reduction on the played speech to obtain the speech noise reduction result of the Bluetooth headset.

Images