OWS earphone multi-scene AI adaptive noise reduction method and system

Abstract

The invention relates to the technical field of receiver ports, and discloses an OWS earphone multi-scene AI adaptive noise reduction method and system, and the system comprises an OWS earphone environment noise detection module, an OWS earphone noise reduction algorithm screening module, and an OWS earphone noise reduction control module. Scientific analysis of the ambient noise level of the OWS earphone is carried out according to the ambient sound frequency preprocessing parameters of the OWS earphone in combination with numerical analysis and standard set standard frequency intervals of different levels of ambient noise of the OWS earphone, and scientific classification noise reduction processing of the ambient noise of the OWS earphone is realized; on the basis of the OWS earphone environment sound frequency preprocessing parameters, a target OWS earphone AI environment noise reduction processing algorithm library and an artificial intelligence recognition algorithm, accurate matching of an optimal AI environment noise reduction processing algorithm is carried out on OWS earphone noise reduction, and the optimal AI environment noise reduction processing algorithm is intelligently screened on the basis of the OWS earphone surrounding environment noise level. And the noise reduction effect of the OWS earphone is improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of receiver ports, in particular to an OWS earphone multi-scene AI adaptive noise reduction method and system. BACKGROUND

[0002] An OWS earphone is a wearable audio device based on an open structure, and natural integration of environmental sound and audio content is achieved through a non-ear design. OWS earphone noise reduction refers to reducing noise by using a certain method. There are two kinds of noise reduction earphones, namely active noise reduction earphones and passive noise reduction earphones. The active noise reduction function is to neutralize noise by generating an opposite sound wave equal to external noise, thereby achieving the effect of noise reduction. Active noise reduction earphones have noise reduction circuits that counteract external noise, and most of them use a large-sized head-mounted design to block external noise by using earplug cotton and earphone housings. Passive noise reduction earphones mainly block external noise by forming a closed space around the ear or using soundproof materials such as silicone earplugs. OWS earphone noise reduction usually uses AI noise reduction algorithms to reduce environmental noise and improve OWS earphone audio quality. However, current OWS earphone noise reduction processing cannot adaptively match the optimal AI environmental noise reduction processing algorithm based on the environmental noise level, which reduces the quality and applicability of OWS earphone noise reduction processing.

[0003] A noise reduction method for earphones is disclosed in Chinese patent CN110430500B, which determines the number of microphones by obtaining the current noise reduction mode, controls the microphones corresponding to the number of microphones to collect environmental noise, and reduces the environmental noise collected by the microphones corresponding to the number of microphones. The above technical solution cannot adaptively match the optimal environmental noise reduction scheme based on the environmental noise level, which reduces the effect of earphone noise reduction. SUMMARY

[0004] (I) Technical problems to be solved

[0005] To solve the above-mentioned problem that current OWS earphone noise reduction processing cannot adaptively match the optimal AI environmental noise reduction processing algorithm based on the environmental noise level, which reduces the quality and applicability of OWS earphone noise reduction processing, the purpose of the above-mentioned self-identification of OWS earphone environmental noise, scientific analysis of OWS earphone environmental noise level, accurate selection of AI environmental noise reduction processing algorithm library required by OWS earphone, intelligent selection of optimal AI environmental noise reduction processing algorithm, and adaptive execution of OWS earphone noise reduction operation is achieved.

[0006] (II) Technical solutions

[0007] The application is implemented by the following technical solutions: an OWS earphone multi-scene AI adaptive noise reduction method, the method comprising the following steps:

[0008] S1, collect OWS earphone ambient sound frequency data;

[0009] S2, OWS earphone ambient sound frequency parameter preprocessing according to the OWS earphone ambient sound frequency data collected, generating OWS earphone ambient sound frequency preprocessing data;

[0010] S3, based on the OWS earphone ambient sound frequency preprocessing data and OWS earphone ambient noise standard frequency interval, the noise recognition processing of OWS earphone surrounding environment is carried out, and OWS earphone ambient noise recognition data is generated. When it is not noise, the OWS earphone noise reduction operation is directly ended;

[0011] S4, when it is noise, according to the OWS earphone ambient sound frequency preprocessing data and OWS earphone different grade ambient noise standard frequency interval, the noise level analysis processing of OWS earphone surrounding environment is carried out, and OWS earphone ambient noise level analysis data is generated;

[0012] S5, according to the OWS earphone ambient noise level analysis data and OWS earphone AI ambient noise reduction processing algorithm library, the AI ambient noise reduction processing algorithm library required for OWS earphone noise reduction is searched and processed, and the target OWS earphone AI ambient noise reduction processing algorithm library is generated;

[0013] S6, based on the OWS earphone ambient sound frequency preprocessing data and the target OWS earphone AI ambient noise reduction processing algorithm library, the optimal AI ambient noise reduction processing algorithm analysis processing required for OWS earphone noise reduction is carried out, and OWS earphone noise reduction optimal noise reduction algorithm analysis data is generated;

[0014] S7, OWS earphone noise reduction collection data is constructed and OWS earphone ambient noise reduction processing operation is executed.

[0015] Preferably, the operation steps of collecting OWS earphone ambient sound frequency data are as follows:

[0016] S11, the sound frequency information of OWS earphone surrounding environment is collected in real time through the built-in microphone of earphone, and OWS earphone ambient sound frequency data is generated , wherein The unit is hertz.

[0017] Preferably, the operation steps of OWS earphone ambient sound frequency parameter preprocessing according to the OWS earphone ambient sound frequency data collected to generate OWS earphone ambient sound frequency preprocessing data are as follows:

[0018] S21, the moving average method is used to process the OWS earphone ambient sound frequency data Data noise preprocessing of OWS earphone ambient sound frequency parameters is performed, and OWS earphone ambient sound frequency preprocessing data is generated .

[0019] Preferably, noise identification processing of the OWS earphone surrounding environment is performed based on the OWS earphone ambient sound frequency preprocessing data and the OWS earphone ambient noise standard frequency interval, and OWS earphone ambient noise identification data is generated. When it is not noise, the operation steps of the OWS earphone noise reduction operation are as follows:

[0020] S31, establishing an OWS earphone ambient noise standard frequency interval , wherein and respectively represent the minimum value and the maximum value of the OWS earphone ambient noise standard frequency in the OWS earphone ambient noise standard frequency interval , and and are both in hertz;

[0021] S32, using the Boyer-Moore search algorithm to compare the OWS earphone ambient sound frequency preprocessing data with the minimum value and the maximum value of the OWS earphone ambient noise standard frequency in the OWS earphone ambient noise standard frequency interval , and generating OWS earphone ambient noise identification data according to the frequency value comparison result ;

[0022] When ∈ , it means that the OWS earphone surrounding environment sound is noise, and the OWS earphone ambient noise identification data is output as noise;

[0023] When ∉ , it means that the OWS earphone surrounding environment sound is not noise, and the OWS earphone ambient noise identification data is output as non-noise, and the OWS earphone noise reduction operation is directly ended.

[0024] Preferably, when it is noise, noise level analysis processing of the OWS earphone surrounding environment is performed according to the OWS earphone ambient sound frequency preprocessing data and the OWS earphone different level ambient noise standard frequency interval, and OWS earphone ambient noise level analysis data is generated. The operation steps are as follows:

[0025] S41, when the OWS earphone environment noise recognition data is noise, establish OWS earphone different level environment noise standard frequency interval set , wherein represents the OWS earphone first level environment noise standard frequency interval, represents the OWS earphone second level environment noise standard frequency interval, represents the OWS earphone third level environment noise standard frequency interval; , wherein and respectively represent the OWS earphone first level environment noise standard frequency minimum value and the OWS earphone first level environment noise standard frequency maximum value in the OWS earphone first level environment noise standard frequency interval ; , wherein and respectively represent the OWS earphone second level environment noise standard frequency minimum value and the OWS earphone second level environment noise standard frequency maximum value in the OWS earphone second level environment noise standard frequency interval ; , wherein and respectively represent the OWS earphone third level environment noise standard frequency minimum value and the OWS earphone third level environment noise standard frequency maximum value in the OWS earphone third level environment noise standard frequency interval , wherein , , , , , The unit of each of the above is hertz;

[0026] S42, the OWS earphone environment sound frequency preprocessing data respectively with the OWS earphone different level environment noise standard frequency interval set inside the OWS earphone first level environment noise standard frequency interval The OWS earphone first level environment noise standard frequency minimum value , the OWS earphone first level environment noise standard frequency maximum value , the OWS earphone second level environment noise standard frequency interval The OWS earphone second level environment noise standard frequency minimum value , the OWS earphone second level environment noise standard frequency maximum value , and the OWS earphone third level environment noise standard frequency interval The minimum value of the third-grade ambient noise standard frequency of the OWS earphone The maximum value of the third-grade ambient noise standard frequency of the OWS earphone The frequency data comparison is performed to search for the first-grade ambient noise standard frequency interval of the OWS earphone The first-grade ambient noise standard frequency interval of the OWS earphone The second-grade ambient noise standard frequency interval of the OWS earphone The third-grade ambient noise standard frequency interval of the OWS earphone The corresponding OWS earphone ambient noise level text information, and the OWS earphone ambient noise level analysis data generated through data identification .

[0027] Preferably, according to the OWS earphone ambient noise level analysis data and the OWS earphone AI ambient noise reduction processing algorithm library required for OWS earphone noise reduction, the operation steps of generating the target OWS earphone AI ambient noise reduction processing algorithm library are as follows:

[0028] S51, establishing an OWS earphone AI ambient noise reduction processing algorithm library set , wherein represents the OWS earphone first-grade ambient noise AI ambient noise reduction processing algorithm library, represents the OWS earphone second-grade ambient noise AI ambient noise reduction processing algorithm library, represents the OWS earphone third-grade ambient noise AI ambient noise reduction processing algorithm library; wherein the OWS earphone AI ambient noise reduction processing algorithm library represents the data storage library of the optimal AI ambient noise reduction processing algorithm set for different grades of ambient noise of the OWS earphone; the AI ambient noise reduction processing algorithm includes RNNoise, CRNN, DNN, DTLN and FullSubNet;

[0029] S52, using a K-D tree nearest neighbor search algorithm to search for the OWS earphone ambient noise level analysis data in the OWS earphone AI ambient noise reduction processing algorithm library set , the OWS earphone first-grade ambient noise AI ambient noise reduction processing algorithm library , the OWS earphone second-grade ambient noise AI ambient noise reduction processing algorithm library , and the OWS earphone third-grade ambient noise AI ambient noise reduction processing algorithm library The OWS earphone environmental noise level character matching is performed to search for the OWS earphone environmental noise level analysis data The matched OWS earphone first-level environmental noise AI environmental noise reduction processing algorithm library Or the OWS earphone second-level environmental noise AI environmental noise reduction processing algorithm library Or the WS earphone third-level environmental noise AI environmental noise reduction processing algorithm library And the target OWS earphone AI environmental noise reduction processing algorithm library is constructed through data identification , ; wherein The target OWS earphone AI environmental noise reduction processing algorithm library The target OWS earphone AI environmental noise reduction processing algorithm corresponding to the The target OWS earphone AI environmental noise reduction processing algorithm corresponding to the The maximum value of the environmental noise frequency value type quantity; the target OWS earphone AI environmental noise reduction processing algorithm indicates an optimal AI environmental noise reduction processing execution program for the environmental noise level information and environmental noise frequency parameter setting of the environment in which the OWS earphone is located.

[0030] Preferably, the OWS earphone noise reduction optimal noise reduction algorithm analysis data is generated through the following operation steps based on the OWS earphone environmental sound frequency preprocessing data and the target OWS earphone AI environmental noise reduction processing algorithm library:

[0031] S61, obtaining the OWS earphone environmental sound frequency preprocessing data , the target OWS earphone AI environmental noise reduction processing algorithm library ;

[0032] S62, comparing the OWS earphone environmental sound frequency preprocessing data With the target OWS earphone AI environmental noise reduction processing algorithm in the target OWS earphone AI environmental noise reduction processing algorithm library The target OWS earphone AI environmental noise reduction processing algorithm The target OWS earphone AI environmental noise reduction processing algorithm corresponding to the OWS earphone environmental sound frequency preprocessing data The target OWS earphone AI environmental noise reduction processing algorithm And construct the OWS earphone noise reduction optimal noise reduction algorithm analysis data The specific operation steps of constructing the OWS earphone noise reduction optimal noise reduction algorithm analysis data are as follows:

[0033] S621, initialization, update the maximum number of iterations T, update the generation of noise reduction algorithm to identify the position of the white shark population, the noise reduction algorithm to identify the position of the white shark population calculation formula as follows: wherein denotes the noise reduction algorithm to identify the white shark individual In the spatial dimension is The target OWS earphone AI environmental noise reduction processing algorithm library in the search space, and denote the upper limit and lower limit of the search space of the target OWS earphone AI environmental noise reduction processing algorithm library , respectively, denotes a random number between the values ;

[0034] S622, speed update stage, noise reduction algorithm to identify the white shark according to the prey movement to perceive its position, and update its own speed, in the search space of the target OWS earphone AI environmental noise reduction processing algorithm library The search space of the target OWS earphone AI environmental noise reduction processing algorithm library matched with the OWS earphone environmental sound frequency preprocessing data , the speed of its own update calculation formula as follows: wherein denotes the noise reduction algorithm to identify the white shark individual After iterations in the search space of the target OWS earphone AI environmental noise reduction processing algorithm library , the speed of its own update calculation formula as follows: denotes the noise reduction algorithm to identify the white shark individual After iterations in the search space of the target OWS earphone AI environmental noise reduction processing algorithm library , the speed of its own update calculation formula as follows: denotes the optimal position of the noise reduction algorithm to identify the white shark individual in the search space of the target OWS earphone AI environmental noise reduction processing algorithm library after iterations, denotes the noise reduction algorithm to identify the white shark individual After iterations in the search space of the target OWS earphone AI environmental noise reduction processing algorithm library , the position of the noise reduction algorithm to identify the white shark individual; denotes the noise reduction algorithm to identify the white shark individual After iterations in the search space of the target OWS earphone AI environmental noise reduction processing algorithm library the search space of the target OWS headset AI ambient noise reduction processing algorithm library corresponding to the speed denote the contraction coefficient of the algorithm, and denote the control coefficient of and , respectively, and both denote a random number between the values

[0035] S623, position updating stage, the noise reduction algorithm identifies white shark updates its position in the search space of the optimal prey by moving towards the optimal prey, the noise reduction algorithm identifies white shark in the target OWS headset AI ambient noise reduction processing algorithm library searches for the target OWS headset AI ambient noise reduction processing algorithm that optimally matches or sub-optimally matches the OWS headset ambient sound frequency preprocessing data prey, the noise reduction algorithm identifies white shark position updating calculation formula is as follows: , wherein denotes the noise reduction algorithm identifies white shark individual searching for prey in the position updating stage after iterations in the search space of the target OWS headset AI ambient noise reduction processing algorithm library , denotes a bitwise operator, is a logical vector, and are both binary vectors, denotes the attractive force coefficient of the noise reduction algorithm identifies white shark approaching the prey, denotes the wave frequency of the movement product of the noise reduction algorithm identifies white shark;

[0036] The noise reduction algorithm identifies white shark moves towards the optimal noise reduction algorithm identifies white shark position to approach the optimal position of the prey, and the noise reduction algorithm identifies white shark position updating calculation formula is as follows:

[0037] , wherein denotes the noise reduction algorithm identifies white shark individual searching for the optimal noise reduction algorithm identifies white shark position in the position updating stage after iterations in the search space of the target OWS headset AI ambient noise reduction processing algorithm library , , , both denote a value ​​​Random number between denotes the distance between the white shark and the prey identified by the noise reduction algorithm, in the search space of the target OWS earphone AI environmental noise reduction algorithm library that matches the OWS earphone environmental sound frequency preprocessing data ; denotes the sign return function, denotes the olfactory and visual parameters of the white shark approaching the optimal prey identified by the noise reduction algorithm;

[0038] S624, fish behavior stage, the white shark population identified by the noise reduction algorithm preserves the optimal noise reduction algorithm identified by the white shark position in the search space of the target OWS earphone AI environmental noise reduction algorithm library update other noise reduction algorithm identified white shark individual position to obtain the target OWS earphone AI environmental noise reduction algorithm library that matches the OWS earphone environmental sound frequency preprocessing data ; , where denotes the fish behavior stage update of other noise reduction algorithm identified white shark individual after iterations in the search space of the target OWS earphone AI environmental noise reduction algorithm library , the target OWS earphone AI environmental noise reduction algorithm library that matches the OWS earphone environmental sound frequency preprocessing data ;

[0039] S625, when the maximum number of iterations is satisfied, the target OWS earphone AI environmental noise reduction algorithm that matches the OWS earphone environmental sound frequency preprocessing data is output, and the OWS earphone noise reduction optimal noise reduction algorithm analysis data is constructed through data identification. The OWS earphone noise reduction optimal noise reduction algorithm analysis data indicates the optimal AI environmental noise reduction processing algorithm required for OWS earphone noise reduction in the current noise environment of OWS earphone.

[0040] ​​​Preferably, the OWS earphone noise reduction collection data is constructed and the OWS earphone ambient noise reduction processing operation is performed according to the following steps:

[0041] S71, the OWS earphone noise reduction optimal noise reduction algorithm analysis data The OWS earphone noise reduction collection data is constructed through data identification ;

[0042] S72, the OWS earphone management platform according to the OWS earphone noise reduction collection data The corresponding AI ambient noise reduction processing execution program performs OWS earphone ambient noise reduction processing operation.

[0043] An OWS earphone multi-scene AI adaptive noise reduction system for realizing the OWS earphone multi-scene AI adaptive noise reduction method, the system comprising an OWS earphone ambient noise detection module, an OWS earphone noise reduction algorithm screening module, and an OWS earphone noise reduction control module.

[0044] The OWS earphone ambient noise detection module comprises an OWS earphone ambient sound frequency acquisition unit, an OWS earphone ambient sound frequency preprocessing unit, an OWS earphone ambient noise standard frequency interval storage unit, and an OWS earphone ambient noise recognition unit.

[0045] The OWS earphone ambient sound frequency acquisition unit acquires OWS earphone ambient sound frequency data through an earphone built-in microphone; the OWS earphone ambient sound frequency preprocessing unit performs OWS earphone ambient sound frequency parameter preprocessing according to the OWS earphone ambient sound frequency data to generate OWS earphone ambient sound frequency preprocessing data; the OWS earphone ambient noise standard frequency interval storage unit is used for storing OWS earphone ambient noise standard frequency intervals; and the OWS earphone ambient noise recognition unit performs noise recognition processing of the OWS earphone ambient noise based on the OWS earphone ambient sound frequency preprocessing data and the OWS earphone ambient noise standard frequency intervals to generate OWS earphone ambient noise recognition data.

[0046] The OWS earphone noise reduction algorithm screening module comprises different levels of OWS earphone ambient noise standard frequency interval storage units, an OWS earphone ambient noise level analysis unit, an OWS earphone AI ambient noise reduction processing algorithm library storage unit, a target OWS earphone AI ambient noise reduction processing algorithm library search unit, and an OWS earphone optimal AI ambient noise reduction processing algorithm analysis unit.

[0047] The different grade OWS earphone ambient noise standard frequency interval storage unit is used for storing different grade OWS earphone ambient noise standard frequency intervals; the OWS earphone ambient noise grade analysis unit performs noise grade analysis processing on the ambient environment of the OWS earphone according to the OWS earphone ambient sound frequency preprocessing data and the different grade OWS earphone ambient noise standard frequency intervals, and generates OWS earphone ambient noise grade analysis data; the OWS earphone AI ambient noise reduction processing algorithm library storage unit is used for storing an OWS earphone AI ambient noise reduction processing algorithm library; the target OWS earphone AI ambient noise reduction processing algorithm library searching unit searches the AI ambient noise reduction processing algorithm library required for OWS earphone noise reduction according to the OWS earphone ambient noise grade analysis data and the OWS earphone AI ambient noise reduction processing algorithm library, and generates a target OWS earphone AI ambient noise reduction processing algorithm library; and the OWS earphone optimal AI ambient noise reduction processing algorithm analysis unit analyzes the optimal AI ambient noise reduction processing algorithm required for OWS earphone noise reduction based on the OWS earphone ambient sound frequency preprocessing data and the target OWS earphone AI ambient noise reduction processing algorithm library, and generates OWS earphone noise reduction optimal noise reduction algorithm analysis data.

[0048] The OWS earphone noise reduction control module comprises an OWS earphone noise reduction information collection unit and an OWS earphone noise reduction operation execution unit.

[0049] The OWS earphone noise reduction information collection unit constructs OWS earphone noise reduction collection data based on OWS earphone noise reduction optimal noise reduction algorithm analysis result information; and the OWS earphone noise reduction operation execution unit executes OWS earphone ambient noise reduction processing operation according to the AI ambient noise reduction processing execution program corresponding to the OWS earphone noise reduction collection data.

[0050] (Three) beneficial effects

[0051] The application provides an OWS earphone multi-scene AI adaptive noise reduction method and system.

[0052] I. The OWS earphone environment sound frequency information is dynamically collected through the built-in microphone of the earphone, and the frequency parameter text noise preprocessing is combined to improve the OWS earphone ambient noise reduction quality; based on the OWS earphone ambient sound frequency preprocessing parameter, the intelligent search algorithm and the scientifically stored OWS earphone ambient noise standard frequency interval are combined to perform OWS earphone ambient environment noise digital efficient identification, so that the OWS earphone can independently and accurately detect the ambient noise and improve the response rate of the OWS earphone noise reduction.

[0053] II. Through the OWS earphone ambient sound frequency preprocessing parameter combined with numerical analysis and standard setting OWS earphone different level ambient noise standard frequency interval, the OWS earphone surrounding environment noise level scientific analysis is realized, the OWS earphone surrounding environment noise scientific classification noise reduction processing is realized, and the precision of OWS earphone noise reduction processing is improved;According to the OWS earphone ambient noise level analysis parameter combined with intelligent search algorithm and OWS earphone AI ambient noise reduction processing algorithm library, the OWS earphone noise reduction AI ambient noise reduction processing algorithm library intelligent search is realized, and the AI ambient noise reduction processing algorithm library based on OWS earphone surrounding environment noise level is realized;Based on OWS earphone ambient sound frequency preprocessing parameter, target OWS earphone AI ambient noise reduction processing algorithm library combined with artificial intelligence recognition algorithm, the OWS earphone noise reduction optimal AI ambient noise reduction processing algorithm is accurately matched, the optimal AI ambient noise reduction processing algorithm based on OWS earphone surrounding environment noise level is realized, and the effect of OWS earphone noise reduction is improved.

[0054] III. Through the OWS earphone noise reduction optimal noise reduction algorithm analysis result information combined with data processing, the OWS earphone noise reduction collection information is accurately constructed, and the OWS earphone environment noise reduction processing operation is accurately executed by combining the WS earphone management platform, the OWS earphone noise reduction efficient and reliable dynamic response is realized, and the quality and applicability of OWS earphone noise reduction are improved. BRIEF DESCRIPTION OF DRAWINGS

[0055] Fig. 1 A module schematic diagram of an OWS earphone multi-scene AI adaptive noise reduction system provided by the application;

[0056] Fig. 2 A flowchart of an OWS earphone multi-scene AI adaptive noise reduction method provided by the application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0058] The implementation of the OWS earphone multi-scene AI adaptive noise reduction method and system is as follows:

[0059] Embodiment 1:

[0060] Please refer to Figs. 1-2 An OWS earphone multi-scene AI adaptive noise reduction method, the method comprising the following steps:

[0061] S1, collect OWS earphone ambient sound frequency data;

[0062] S2, OWS earphone ambient sound frequency parameter preprocessing according to OWS earphone ambient sound frequency data collected, generating OWS earphone ambient sound frequency preprocessing data;

[0063] S3, based on OWS earphone ambient sound frequency preprocessing data and OWS earphone ambient noise standard frequency interval, OWS earphone ambient noise recognition processing is carried out, and OWS earphone ambient noise recognition data is generated. When it is not noise, this OWS earphone noise reduction operation is directly ended;

[0064] S4, when it is noise, OWS earphone ambient noise level analysis processing is carried out according to OWS earphone ambient sound frequency preprocessing data and OWS earphone different grade ambient noise standard frequency interval, and OWS earphone ambient noise level analysis data is generated;

[0065] S5, according to OWS earphone ambient noise level analysis data and OWS earphone AI ambient noise reduction processing algorithm library, OWS earphone AI ambient noise reduction processing algorithm library search processing required for OWS earphone noise reduction is carried out, and target OWS earphone AI ambient noise reduction processing algorithm library is generated;

[0066] S6, based on OWS earphone ambient sound frequency preprocessing data and target OWS earphone AI ambient noise reduction processing algorithm library, OWS earphone noise reduction required optimal AI ambient noise reduction processing algorithm analysis processing is carried out, and OWS earphone noise reduction optimal noise reduction algorithm analysis data is generated;

[0067] S7, OWS earphone noise reduction collection data is constructed and OWS earphone ambient noise reduction processing operation is executed.

[0068] Further, please refer to Figs. 1-2 , the operation steps of collecting OWS earphone ambient sound frequency data are as follows:

[0069] S11, the sound frequency information of OWS earphone ambient sound is collected in real time through the built-in microphone of the earphone, and OWS earphone ambient sound frequency data is generated , the unit of which is hertz.

[0070] The operation steps of OWS earphone ambient sound frequency parameter preprocessing according to OWS earphone ambient sound frequency data collected, generating OWS earphone ambient sound frequency preprocessing data are as follows:

[0071] S21, the moving average method is adopted to process OWS earphone ambient sound frequency data ​Data noise preprocessing of OWS earphone ambient sound frequency parameters is performed, and OWS earphone ambient sound frequency preprocessing data is generated .

[0072] Noise identification processing of the OWS earphone ambient environment is performed based on the OWS earphone ambient sound frequency preprocessing data and the OWS earphone ambient noise standard frequency interval, and OWS earphone ambient noise identification data is generated. When it is not noise, the operation steps of the OWS earphone noise reduction operation are as follows:

[0073] S31, establishing an OWS earphone ambient noise standard frequency interval , wherein and respectively represent the minimum value and the maximum value of the OWS earphone ambient noise standard frequency in the OWS earphone ambient noise standard frequency interval , and and are both in hertz;

[0074] S32, using the Boyer-Moore search algorithm to compare the OWS earphone ambient sound frequency preprocessing data with the minimum value and the maximum value of the OWS earphone ambient noise standard frequency in the OWS earphone ambient noise standard frequency interval , and generating OWS earphone ambient noise identification data according to the frequency value comparison result;

[0075] When ∈ , it means that the OWS earphone ambient environment sound is noise, and the OWS earphone ambient noise identification data is output as noise;

[0076] When ∉ , it means that the OWS earphone ambient environment sound is not noise, and the OWS earphone ambient noise identification data is output as non-noise, and the OWS earphone noise reduction operation is directly ended.

[0077] The OWS earphone ambient sound frequency acquisition unit and the OWS earphone ambient sound frequency preprocessing unit cooperate with each other, dynamically acquire OWS earphone ambient sound frequency information by using the microphone built-in the earphone, and improve the OWS earphone ambient noise reduction quality by combining the frequency parameter text noise preprocessing.

[0078] Further, please refer to Figs. 1-2 When it is noise, the noise level of the OWS earphone surrounding environment is analyzed and processed according to the OWS earphone ambient sound frequency preprocessing data and the OWS earphone different grade ambient noise standard frequency interval, and the operation steps of generating the OWS earphone ambient noise level analysis data are as follows:

[0079] S41, when the OWS earphone ambient noise recognition data When it is noise, the noise level of the OWS earphone surrounding environment is analyzed and processed according to the OWS earphone ambient sound frequency preprocessing data and the OWS earphone different grade ambient noise standard frequency interval, and the operation steps of generating the OWS earphone ambient noise level analysis data are as follows: , wherein represents the OWS earphone first grade ambient noise standard frequency interval, represents the OWS earphone second grade ambient noise standard frequency interval, represents the OWS earphone third grade ambient noise standard frequency interval; , wherein and respectively represent the OWS earphone first grade ambient noise standard frequency minimum value and the OWS earphone first grade ambient noise standard frequency maximum value in the OWS earphone first grade ambient noise standard frequency interval . , wherein and respectively represent the OWS earphone second grade ambient noise standard frequency minimum value and the OWS earphone second grade ambient noise standard frequency maximum value in the OWS earphone second grade ambient noise standard frequency interval . , wherein and respectively represent the OWS earphone third grade ambient noise standard frequency minimum value and the OWS earphone third grade ambient noise standard frequency maximum value in the OWS earphone third grade ambient noise standard frequency interval , wherein , , , , . The unit of is Hertz;

[0080] S42, pre-processing the OWS headset ambient sound frequency data The standard frequency ranges of different levels of environmental noise for OWS headphones are respectively Internal OWS earphones first level environmental noise standard frequency range Minimum frequency of the first-level environmental noise standard for OWS headphones 、Maximum value of the first-level environmental noise standard frequency of OWS headphones , OWS headphones second level environmental noise standard frequency range Minimum frequency of the second level environmental noise standard for OWS headphones 、Maximum frequency of the second level environmental noise standard of OWS headphones , and the third level environmental noise standard frequency range of OWS headphones Minimum frequency of the third level environmental noise standard for OWS headphones 、Maximum value of the third level environmental noise standard frequency of OWS headphones Compare the frequency data and search for the pre-processed data of the ambient sound frequency with the OWS headset Matching OWS headphones first-level environmental noise standard frequency range Or the second level environmental noise standard frequency range of OWS headphones Or the third level environmental noise standard frequency range of OWS headphones The corresponding OWS headset environmental noise level text information, and generate OWS headset environmental noise level analysis data through data identification .

[0081] Based on the OWS headphone ambient noise level analysis data and the OWS headphone AI ambient noise reduction processing algorithm library, the AI ​​ambient noise reduction processing algorithm library required for OWS headphone noise reduction is searched and processed. The steps for generating the target OWS headphone AI ambient noise reduction processing algorithm library are as follows:

[0082] S51. Establish a library of AI noise reduction algorithms for OWS headphones ,in Indicates the OWS headset's first-level ambient noise AI ambient noise reduction processing algorithm library, Indicates the OWS headset second-level environmental noise AI environmental noise reduction processing algorithm library, The OWS earphone third-grade environmental noise AI environmental noise reduction processing algorithm library represents a data repository of optimal AI environmental noise reduction processing algorithms set for different grades of environmental noise in which the OWS earphone is located. The AI environmental noise reduction processing algorithms include RNNoise, CRNN, DNN, DTLN, and FullSubNet.

[0083] S52, using the K-D tree nearest neighbor search algorithm to analyze the OWS earphone environmental noise level data The OWS earphone AI environmental noise reduction processing algorithm library set The OWS earphone first-grade environmental noise AI environmental noise reduction processing algorithm library The OWS earphone second-grade environmental noise AI environmental noise reduction processing algorithm library The OWS earphone third-grade environmental noise AI environmental noise reduction processing algorithm library The OWS earphone environmental noise level character matching is performed, and the OWS earphone environmental noise level analysis data The OWS earphone first-grade environmental noise AI environmental noise reduction processing algorithm library The OWS earphone second-grade environmental noise AI environmental noise reduction processing algorithm library The OWS earphone third-grade environmental noise AI environmental noise reduction processing algorithm library The target OWS earphone AI environmental noise reduction processing algorithm library is constructed through data identification , ; wherein The target OWS earphone AI environmental noise reduction processing algorithm library The target OWS earphone AI environmental noise reduction processing algorithm corresponding to the The target OWS earphone AI environmental noise reduction processing algorithm corresponding to the The maximum value of the number of environmental noise frequency value types; the target OWS earphone AI environmental noise reduction processing algorithm represents an optimal AI environmental noise reduction processing execution program set for the environmental noise level information and environmental noise frequency parameters of the environment in which the OWS earphone is located.

[0084] Based on the OWS earphone environmental sound frequency preprocessing data and the target OWS earphone AI environmental noise reduction processing algorithm library, the optimal AI environmental noise reduction processing algorithm analysis and processing required for OWS earphone noise reduction are performed to generate the OWS earphone noise reduction optimal noise reduction algorithm analysis data. The operation steps are as follows:

[0085] S61, obtaining OWS earphone environmental sound frequency preprocessing data , target OWS earphone AI ambient noise reduction processing algorithm library ;

[0086] S62, OWS earphone environmental sound frequency preprocessing data and target OWS earphone AI ambient noise reduction processing algorithm library target OWS earphone AI ambient noise reduction processing algorithm frequency value comparison, search OWS earphone environmental sound frequency preprocessing data corresponding target OWS earphone AI ambient noise reduction processing algorithm , and construct OWS earphone noise reduction optimal noise reduction algorithm analysis data ; The specific operation steps of constructing OWS earphone noise reduction optimal noise reduction algorithm analysis data are as follows:

[0087] S621, initialization, update the maximum iteration number T, update the generated noise reduction algorithm identification white shark population individual position, and the noise reduction algorithm identification white shark population individual position calculation formula is as follows: , wherein denotes the noise reduction algorithm identification white shark individual in the search space of the target OWS earphone AI ambient noise reduction processing algorithm library with spatial dimension , and and respectively represent the upper limit and the lower limit of the search space of the target OWS earphone AI ambient noise reduction processing algorithm library , and denotes a random number between the values ;

[0088] S622, speed update stage, noise reduction algorithm identification white shark according to the movement of prey to perceive its position, and update its own speed, search the target OWS earphone AI ambient noise reduction processing algorithm matching OWS earphone environmental sound frequency preprocessing data in the search space of the target OWS earphone AI ambient noise reduction processing algorithm library , and the own speed update calculation formula is as follows: , wherein denotes the noise reduction algorithm identification white shark individual in the search space of the target OWS earphone AI ambient noise reduction processing algorithm library after iterations, denotes the noise reduction algorithm identification white shark individual in the search space of the target OWS earphone AI ambient noise reduction processing algorithm library After iterations, the target OWS headset AI environmental noise reduction processing algorithm library The speed in the search space, Indicates After iterations, the noise reduction algorithm identifies the white shark individual in the target OWS headset AI environmental noise reduction processing algorithm library The optimal position in the search space of Represents the noise reduction algorithm to identify white shark individuals exist After iterations, the noise reduction algorithm identifies the white shark individual in the target OWS headset AI environmental noise reduction processing algorithm library The position in the search space of Represents the noise reduction algorithm to identify white shark individuals exist After iterations, the target OWS headset AI environmental noise reduction processing algorithm library In the search space of speed corresponding positions; represents the algorithm shrinkage coefficient, and Respectively and The control coefficient, and Both represent values A random number between

[0089] S623, position update phase, the noise reduction algorithm identifies the white shark and updates its position in the target by moving towards the optimal prey OWS headset AI environmental noise reduction processing algorithm library The position in the search space of the target OWS headset AI environmental noise reduction processing algorithm library Search the search space to find the pre-processed data of the OWS headset environment sound frequency Target OWS headset AI ambient noise reduction processing algorithm for optimal or suboptimal matching The prey,noise reduction algorithm identifies the white shark's position and updates the calculation formula as follows: ,in Representation of the denoising algorithm for searching prey during the position update phase to identify individual white sharks exist After iterations, the target OWS headset AI environmental noise reduction processing algorithm library The position in the search space, Represents a bitwise operator, is a logical vector, and are binary vectors, represents the attraction coefficient of the noise reduction algorithm to identify the white shark approaching prey, Indicates that the noise reduction algorithm identifies the wave frequency of the White Shark sports product;

[0090] The noise reduction algorithm identifies the white shark by moving towards the optimal position of the white shark to get closer to the optimal position of the prey. The calculation formula for the white shark position update is as follows:

[0091] ,in Represents the search for the optimal noise reduction algorithm in the position update phase to identify the white shark position and the noise reduction algorithm to identify the white shark individual. exist After iterations, the target OWS headset AI environmental noise reduction processing algorithm library The position in the search space, 、 、 Both represent values A random number between Indicates that the noise reduction algorithm identifies the distance between the white shark and its prey, and the target OWS headset AI ambient noise reduction processing algorithm library Search the search space to find the pre-processed data of the OWS headset environment sound frequency Matching target OWS headset AI ambient noise reduction processing algorithm ; Indicates a sign-returning function. It indicates that the noise reduction algorithm identifies the olfactory and visual parameters of white sharks close to the best prey;

[0092] S624, fish behavior stage, the noise reduction algorithm identifies the white shark population through feeding behavior to retain the position update stage in the target OWS headset AI environmental noise reduction processing algorithm library The optimal noise reduction algorithm in the search space identifies the white shark position, and updates other noise reduction algorithms to identify the individual white shark positions based on the optimal noise reduction algorithm to obtain the target OWS headset AI environmental noise reduction processing algorithm library. Search the search space to find the pre-processed data of the OWS headset environment sound frequency Matching target OWS headset AI ambient noise reduction processing algorithm ,The noise reduction algorithm identifies the individual position of white sharks and updates the calculation formula as follows: ,in Indicates the stage of fish school behavior to update other noise reduction algorithms to identify white shark individuals exist After iterations, the target OWS headset AI environmental noise reduction processing algorithm library The position in the search space of the target OWS headset AI environmental noise reduction processing algorithm library searching out the OWS earphone ambient sound frequency preprocessing data matching the target OWS earphone AI ambient noise reduction processing algorithm from the search space the target OWS earphone AI ambient noise reduction processing algorithm ;

[0093] S625, when the maximum number of iterations is met, output the OWS earphone ambient sound frequency preprocessing data the target OWS earphone AI ambient noise reduction processing algorithm , and the OWS earphone noise reduction optimal noise reduction algorithm analysis data is constructed through data identification The OWS earphone noise reduction optimal noise reduction algorithm analysis data indicates the optimal AI ambient noise reduction processing algorithm required by the OWS earphone in the current noise environment.

[0094] Through the OWS earphone ambient noise level analysis unit, the OWS earphone ambient sound frequency preprocessing parameters are combined with numerical analysis and the OWS earphone different level ambient noise standard frequency interval set by the standard to scientifically analyze the OWS earphone ambient noise level, realize the scientific classification noise reduction processing of the OWS earphone ambient noise, and improve the fineness of the OWS earphone noise reduction processing; the target OWS earphone AI ambient noise reduction processing algorithm library search unit, according to the OWS earphone ambient noise level analysis parameters, combines the intelligent search algorithm with the OWS earphone AI ambient noise reduction processing algorithm library to intelligently search the AI ambient noise reduction processing algorithm library required by the OWS earphone noise reduction, realizes the scientific construction of the AI ambient noise reduction processing algorithm library based on the OWS earphone ambient noise level; the OWS earphone optimal AI ambient noise reduction processing algorithm analysis unit, based on the OWS earphone ambient sound frequency preprocessing parameters, the target OWS earphone AI ambient noise reduction processing algorithm library, and the artificial intelligence recognition algorithm, accurately matches the optimal AI ambient noise reduction processing algorithm for the OWS earphone noise reduction, realizes the intelligent selection of the optimal AI ambient noise reduction processing algorithm based on the OWS earphone ambient noise level, and improves the effect of the OWS earphone noise reduction.

[0095] Further, please refer to Figs. 1-2 , the operation steps of constructing the OWS earphone noise reduction collection data and performing the OWS earphone ambient noise reduction processing operation are as follows:

[0096] S71, the OWS earphone noise reduction optimal noise reduction algorithm analysis data The OWS earphone noise reduction collection data is constructed through data identification ;

[0097] S72, the OWS earphone management platform, according to the OWS earphone noise reduction collection data The corresponding AI ambient noise reduction processing program executes the OWS earphone ambient noise reduction processing operation.

[0098] Through the cooperation of the OWS earphone noise reduction information collection unit and the OWS earphone noise reduction operation execution unit, the OWS earphone noise reduction collection information is accurately constructed based on the OWS earphone noise reduction optimal noise reduction algorithm analysis result information combined with data processing, and the OWS earphone ambient noise reduction processing operation is accurately executed by the WS earphone management platform, realizing efficient and reliable dynamic response of OWS earphone noise reduction, and improving the quality and applicability of OWS earphone noise reduction.

[0099] Embodiment 2:

[0100] Please refer to Figs. 1-2 An OWS earphone multi-scene AI adaptive noise reduction system for realizing an OWS earphone multi-scene AI adaptive noise reduction method, the system comprising an OWS earphone ambient noise detection module, an OWS earphone noise reduction algorithm screening module, and an OWS earphone noise reduction control module.

[0101] The OWS earphone ambient noise detection module comprises an OWS earphone ambient sound frequency acquisition unit, an OWS earphone ambient sound frequency preprocessing unit, an OWS earphone ambient noise standard frequency interval storage unit, and an OWS earphone ambient noise recognition unit.

[0102] The OWS earphone ambient sound frequency acquisition unit acquires OWS earphone ambient sound frequency data through the built-in microphone of the earphone; the OWS earphone ambient sound frequency preprocessing unit performs OWS earphone ambient sound frequency parameter preprocessing based on the OWS earphone ambient sound frequency data collected, and generates OWS earphone ambient sound frequency preprocessing data; the OWS earphone ambient noise standard frequency interval storage unit is used to store OWS earphone ambient noise standard frequency intervals; and the OWS earphone ambient noise recognition unit performs noise recognition processing of the OWS earphone ambient noise based on the OWS earphone ambient sound frequency preprocessing data and the OWS earphone ambient noise standard frequency interval, and generates OWS earphone ambient noise recognition data.

[0103] The OWS earphone noise reduction algorithm screening module comprises different levels of OWS earphone ambient noise standard frequency interval storage units, an OWS earphone ambient noise level analysis unit, an OWS earphone AI ambient noise reduction processing algorithm library storage unit, a target OWS earphone AI ambient noise reduction processing algorithm library search unit, and an OWS earphone optimal AI ambient noise reduction processing algorithm analysis unit.

[0104] A storage unit for standard frequency intervals of environmental noise of different levels for OWS headphones is used to store standard frequency intervals of environmental noise of different levels for OWS headphones; an OWS headphone environmental noise level analysis unit performs noise level analysis on the environment around the OWS headphones based on the pre-processed data of the environmental sound frequency of the OWS headphones and the standard frequency intervals of environmental noise of different levels for the OWS headphones, and generates analysis data of the environmental noise level of the OWS headphones; a storage unit for the AI ​​environmental noise reduction processing algorithm library of the OWS headphones is used to store the AI ​​environmental noise reduction processing algorithm library of the OWS headphones; a search unit for the AI ​​environmental noise reduction processing algorithm library of the target OWS headphones is used to search for the AI ​​environmental noise reduction processing algorithm library required for noise reduction of the OWS headphones based on the analysis data of the environmental noise level of the OWS headphones and the AI ​​environmental noise reduction processing algorithm library of the OWS headphones, and generates the target OWS headphones AI environmental noise reduction processing algorithm library; an analysis unit for the optimal AI environmental noise reduction processing algorithm required for noise reduction of the OWS headphones is used to analyze the optimal AI environmental noise reduction processing algorithm required for noise reduction of the OWS headphones based on the pre-processed data of the environmental sound frequency of the OWS headphones and the target OWS headphones AI environmental noise reduction processing algorithm library, and generates analysis data of the optimal noise reduction algorithm for noise reduction of the OWS headphones;

[0105] The OWS headphone noise reduction control module includes an OWS headphone noise reduction information collection unit and an OWS headphone noise reduction operation execution unit;

[0106] The OWS headphone noise reduction information collection unit constructs the OWS headphone noise reduction collection data based on the analysis result information of the OWS headphone noise reduction optimal noise reduction algorithm; the OWS headphone noise reduction operation execution unit, the OWS headphone management platform executes the OWS headphone environmental noise reduction processing operation according to the AI ​​environmental noise reduction processing execution program corresponding to the OWS headphone noise reduction collection data.

[0107] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An OWS headset multi-scenario AI adaptive noise reduction method, characterized in that: The method comprises the following steps: S1, collect OWS headset ambient sound frequency data; S2. Preprocess the collected OWS headphone ambient sound frequency parameters to generate OWS headphone ambient sound frequency preprocessing data; S3: Perform noise recognition processing on the environment around the OWS headset to generate OWS headset environmental noise recognition data. If the noise is non-noise, the OWS headset noise reduction operation is terminated directly. S4. When the noise level is detected, the noise level of the surrounding environment of the OWS headset is analyzed based on the pre-processed frequency data of the OWS headset environment sound and the standard frequency intervals of different levels of ambient noise of the OWS headset to generate OWS headset environment noise level analysis data. S5. Search and process the AI ​​ambient noise reduction processing algorithm library required for OWS headphone noise reduction to generate the target OWS headphone AI ambient noise reduction processing algorithm library; S6. Analyze and process the optimal AI environmental noise reduction algorithm required for OWS headphone noise reduction based on the OWS headphone ambient sound frequency preprocessing data and the target OWS headphone AI environmental noise reduction algorithm library to generate OWS headphone noise reduction optimal noise reduction algorithm analysis data; S7. Construct OWS headphone noise reduction collection data and execute OWS headphone ambient noise reduction processing operation.

2. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 1, characterized in that: Said S1 comprises the following steps: S11. Use the built-in microphone of the headset to collect the sound frequency information of the environment around the OWS headset in real time, and generate the OWS headset environment sound frequency data ,in The unit is Hertz.

3. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 2, characterized in that: The S2 comprises the following steps: S21, using the moving average method to Perform data noise preprocessing of OWS headset ambient sound frequency parameters and generate OWS headset ambient sound frequency preprocessing data .

4. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 3, characterized in that: The S3 includes the following steps: S31. Establish the standard frequency range of ambient noise for OWS headphones ,in and Respectively represent the standard frequency range of the OWS headphone environmental noise The minimum value of the standard frequency of environmental noise of OWS headphones and the maximum value of the standard frequency of environmental noise of OWS headphones are and The unit of is Hertz; S32, using the Boyer-Moore search algorithm to With the As stated in and stated Perform frequency value comparison and generate OWS headphone environmental noise recognition data based on the frequency value comparison results ; when ∈ , then the output is For noise; when ∉ , then the output is If it is non-noise, the OWS headphone noise reduction operation is ended directly.

5. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 4, characterized in that: The S4 comprises the following steps: S41, when the When there is noise, establish a set of standard frequency ranges for different levels of ambient noise for OWS headphones ,in Indicates the standard frequency range of the first level of environmental noise for OWS headphones. Indicates the second-level environmental noise standard frequency range of OWS headphones. Indicates the third-level environmental noise standard frequency range of OWS headphones; ,in and Respectively represent the The minimum value of the first-level environmental noise standard frequency of OWS headphones and the maximum value of the first-level environmental noise standard frequency of OWS headphones; ,in and Respectively represent the The minimum value of the second-level environmental noise standard frequency of OWS headphones and the maximum value of the second-level environmental noise standard frequency of OWS headphones; ,in and Respectively represent the The minimum frequency value of the third-level environmental noise standard of OWS headphones and the maximum frequency value of the third-level environmental noise standard of OWS headphones, where 、 、 、 、 、 The unit of is Hertz; S42, the Respectively with the Internal As stated in 、 , As stated in 、 , and the As stated in 、 Compare the frequency data and search for the Matching the or the or the The corresponding OWS headset environmental noise level text information, and generate OWS headset environmental noise level analysis data through data identification .

6. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 5, characterized in that: The S5 comprises the following steps: S51. Establish a library of AI noise reduction algorithms for OWS headphones ,in Indicates the OWS headset's first-level ambient noise AI ambient noise reduction processing algorithm library, Indicates the OWS headset second-level environmental noise AI environmental noise reduction processing algorithm library, Indicates the AI ​​noise reduction processing algorithm library for the third level of ambient noise of OWS headphones; S52, using KD tree nearest neighbor search algorithm to With the Internal 、 、 Perform OWS headset environmental noise level character matching and search for the Matching the or the or the , and build the target OWS headset AI environmental noise reduction processing algorithm library through data identification , ;in Represents the target OWS headset AI ambient noise reduction processing algorithm library Middle The target OWS headset AI environmental noise reduction processing algorithm corresponding to the environmental noise frequency value type, Indicates the maximum value of the ambient noise frequency value type.

7. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 6, characterized in that: The S6 comprises the following steps: S61, obtain the 、 ; S62, the With the As stated in Perform frequency value comparison and search for the The corresponding , and build the optimal noise reduction algorithm analysis data for OWS headphones ; Execute and construct the OWS headset noise reduction optimal noise reduction algorithm analysis data The specific steps are as follows: S621, initialization, updating the maximum number of iterations T, and updating the generated noise reduction algorithm to identify the individual positions of the white shark population; S622, speed update stage, the noise reduction algorithm recognizes the white shark's position based on the movement of the prey and updates its own speed. Search the search space for the Matching the ; S623, position update phase, the noise reduction algorithm identifies the white shark and updates itself by moving towards the optimal prey. The position in the search space, Search the search space to find the The best match or suboptimal match Prey, noise reduction algorithm identifies white sharks by moving towards the optimal noise reduction algorithm identifies white shark position to get closer to the optimal prey position; S624, fish school behavior stage, the noise reduction algorithm identifies the white shark population through feeding behavior to retain the position update stage The optimal denoising algorithm in the search space identifies the white shark position, and updates other denoising algorithms to identify the individual white shark positions according to the optimal denoising algorithm to obtain the white shark positions in the search space. Search the search space for the Matching the ; S625. When the maximum number of iterations is met, output the Matching the , and after data identification, the optimal noise reduction algorithm analysis data of OWS headphones is constructed .

8. The OWS headset multi-scenario AI adaptive noise reduction method according to claim 7, characterized in that: The S7 comprises the following steps: S71, the Construct OWS headphone noise reduction collection data through data identification ; S72, OWS headset management platform is based on The corresponding AI environmental noise reduction processing execution program performs the OWS headphone environmental noise reduction processing operation.

9. An OWS headset multi-scenario AI adaptive noise reduction system, configured to implement the OWS headset multi-scenario AI adaptive noise reduction method according to any one of claims 1 to 8, characterized in that: The system includes an OWS headphone ambient noise detection module, an OWS headphone noise reduction algorithm screening module, and an OWS headphone noise reduction control module.

Images