Movatterモバイル変換


[0]ホーム

URL:


CN117202071B - A testing method and system for noise reduction headphones - Google Patents

A testing method and system for noise reduction headphones
Download PDF

Info

Publication number
CN117202071B
CN117202071BCN202311232265.XACN202311232265ACN117202071BCN 117202071 BCN117202071 BCN 117202071BCN 202311232265 ACN202311232265 ACN 202311232265ACN 117202071 BCN117202071 BCN 117202071B
Authority
CN
China
Prior art keywords
noise reduction
noise
data
model
audio
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311232265.XA
Other languages
Chinese (zh)
Other versions
CN117202071A (en
Inventor
林太钦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Jinhaina Industrial Co ltd
Original Assignee
Guangdong Jinhaina Industrial Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong Jinhaina Industrial Co ltdfiledCriticalGuangdong Jinhaina Industrial Co ltd
Priority to CN202311232265.XApriorityCriticalpatent/CN117202071B/en
Publication of CN117202071ApublicationCriticalpatent/CN117202071A/en
Application grantedgrantedCritical
Publication of CN117202071BpublicationCriticalpatent/CN117202071B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

The application relates to the technical field of earphone detection, in particular to a method and a system for testing a noise reduction earphone, wherein the method comprises the following steps: acquiring a scene test type and a noise reduction algorithm model, and acquiring corresponding noise simulation data according to the scene test type; acquiring audio output reference data, inputting noise simulation data into a noise reduction algorithm model one by one according to a scene test type for testing, and respectively comparing the acquired audio noise reduction result with the audio output reference data to obtain a noise reduction test result; correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type; and generating an earphone noise reduction model according to each noise reduction algorithm, and generating model test information according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones. The noise reduction effect of the noise reduction earphone can be improved when the noise reduction earphone is used for a user.

Description

Test method and system of noise reduction earphone
Technical Field
The application relates to the technical field of earphone detection, in particular to a method and a system for testing a noise reduction earphone.
Background
At present, the types of earphones comprise earplug type, head-wearing type, bone conduction type and other types besides the common earphone types, and the earphone with noise reduction function can be provided when a user wears the earphone, so that the external noise is filtered, and the experience of the user when the user uses the earphone is improved.
The existing noise reduction earphone can analyze noise in a use environment in the use process of a user, and the received external noise is calculated through a correlation algorithm, so that the external noise is reduced and transmitted into the user's ear from the earphone, and the use experience of the user is improved. In order to improve the noise reduction effect of the noise reduction earphone in the use process, the noise reduction effect of the noise reduction earphone needs to be tested continuously in the production design process of the noise reduction earphone, however, various noise environments can be met in the use process of a user, and therefore, the test scenes of the noise reduction earphone are enriched continuously in the need.
Disclosure of Invention
In order to improve the noise reduction effect of the noise reduction earphone when the noise reduction earphone is used for a user, the application provides a test method and a test system of the noise reduction earphone.
The first object of the present invention is achieved by the following technical solutions:
a test method of a noise reduction earphone comprises the following steps:
acquiring a scene test type and a noise reduction algorithm model, and acquiring corresponding noise simulation data according to the scene test type;
acquiring audio output reference data, inputting the noise simulation data into the noise reduction algorithm model one by one according to the scene test type for testing, and comparing the acquired audio noise reduction result with the audio output reference data respectively to obtain a noise reduction test result;
correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type;
and generating an earphone noise reduction model according to each noise reduction algorithm, and generating a model test message according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones.
Through adopting above-mentioned technical scheme, when carrying out the test to the noise reduction algorithm in the earphone of making an uproar, through obtaining corresponding noise simulation data according to scene test type, and then can simulate the noise of different scenes, carry out the test of making an uproar, thereby can make an uproar to the noise scene of different scenes pertinently, simultaneously, when carrying out the test, the result of test, noise reduction algorithm generates the earphone model of making an uproar that corresponds promptly, thereby can promote the suitability of noise reduction algorithm and every earphone, and can promote the richness of noise simulation data through the mode of constantly enriching scene test type, and then help promoting the user when making an uproar of making an uproar in follow-up use of earphone, the earphone is to the adaptive degree of environment, with the noise reduction effect of promotion earphone.
The present application may be further configured in a preferred example to: the step of obtaining audio output reference data, which is to input the noise simulation data into the noise reduction algorithm model one by one according to the scene test type for testing, and respectively compare the obtained audio noise reduction result with the audio output reference data to obtain a noise reduction test result, specifically comprising the following steps:
extracting noise characteristic data corresponding to each piece of noise simulation data, and respectively inputting each piece of noise characteristic data into the noise algorithm model;
acquiring the audio noise reduction result corresponding to each noise characteristic data from the noise algorithm model, and extracting each audio noise reduction result to extract the audio noise reduction characteristic data;
and extracting audio reference features from the audio output reference data, and respectively comparing the audio reference features with each audio noise reduction feature to obtain the noise reduction test result corresponding to each noise feature data.
By adopting the technical scheme, because the effect requirement of the user after the noise reduction earphone is used for reducing the noise is unified, when the noise reduction algorithm model is tested, each noise characteristic data is respectively input into the noise reduction algorithm model, so that the noise reduction effect of the noise reduction algorithm model on each noise scene can be tested, namely the audio noise reduction result is compared with the set audio output reference data, the difference between the noise reduction result of the noise reduction algorithm model on each noise simulation data and the audio input reference can be tested, and the subsequent corresponding adjustment is facilitated.
The present application may be further configured in a preferred example to: extracting audio reference features from the audio output reference data, and comparing the audio reference features with each audio noise reduction feature to obtain a noise reduction test result corresponding to each noise feature data, wherein the method specifically comprises the following steps of:
constructing a reference feature vector of the audio reference feature, and constructing a noise reduction feature vector of each audio noise reduction feature;
and calculating comparison distance data of each noise reduction feature vector and the reference feature vector, and taking the comparison distance data as the noise reduction test result.
By adopting the technical scheme, the distance comparison of vectors is carried out by constructing the corresponding feature vectors, so that the difference between each audio noise reduction feature and each audio reference feature can be compared, the actual effect of each audio noise reduction result can be conveniently judged, and the subsequent targeted adjustment of the noise reduction algorithm model to each noise scene is facilitated.
The present application may be further configured in a preferred example to: the noise reduction algorithm model is corrected according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type, and the method specifically comprises the following steps:
sorting each scene test type according to a preset rule, and correcting a noise reduction algorithm model with the first sorting to obtain a first correction result;
inputting noise simulation data corresponding to the second scene test type into the first correction result to obtain correction noise data, and performing difference comparison on the correction noise data and audio output reference data;
if the difference comparison result is within the preset threshold value, the first correction result is used as the corresponding noise reduction algorithm, otherwise, the second noise reduction algorithm model is corrected to obtain a second correction result;
starting from the noise simulation data corresponding to the third scene test type, inputting the corresponding noise simulation data into the corrected results after the noise reduction algorithm model is corrected one by one, and comparing the output results with the audio output reference data in a difference mode so as to screen the corresponding corrected results or correct the noise reduction algorithm model.
By adopting the technical scheme, because the related noise reduction algorithm consumes larger power consumption when the noise reduction earphone is used for noise reduction, especially the use time of the Bluetooth earphone is influenced, the next scene is tested by correcting results obtained by correcting the noise reduction algorithm model one by one, and the applicable correcting results are used as the noise reduction algorithm of the scene at the same time, so that the number of the noise reduction algorithms in the model of the whole noise reduction algorithm can be reduced, the noise reduction algorithm in the noise reduction process is simplified, and the consumption of the noise reduction earphone in use can be reduced.
The present application may be further configured in a preferred example to: the generating the earphone noise reduction model according to each noise reduction algorithm specifically comprises the following steps:
packaging all the noise simulation data to obtain a noise scene recognition model;
and correlating each noise reduction algorithm with corresponding noise simulation data in the noise scene recognition model to obtain an earphone noise reduction model.
By adopting the technical scheme, the noise scene recognition model and the noise reduction algorithm can be separated by adopting the mode of packaging the noise simulation data and correlating the noise scene recognition model with the corresponding noise simulation data, so that scene recognition calculation can be carried out through the terminal connected with the noise reduction earphone through the noise scene recognition model, the calculation of the noise reduction earphone in the noise reduction aspect is further reduced, and the use power consumption of the noise reduction earphone is further reduced.
The second object of the present invention is achieved by the following technical solutions:
a test system for a noise reducing earphone, the test system comprising:
the scene simulation module is used for acquiring a scene test type and a noise reduction algorithm model and acquiring corresponding noise simulation data according to the scene test type;
the algorithm test module is used for acquiring audio output reference data, inputting the noise simulation data into the noise reduction algorithm model one by one according to the scene test type for testing, and comparing the acquired audio noise reduction result with the audio output reference data respectively to obtain a noise reduction test result;
the algorithm correction module is used for respectively correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type;
and the model test module is used for generating an earphone noise reduction model according to each noise reduction algorithm, and generating a model test message according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones.
Through adopting above-mentioned technical scheme, when carrying out the test to the noise reduction algorithm in the earphone of making an uproar, through obtaining corresponding noise simulation data according to scene test type, and then can simulate the noise of different scenes, carry out the test of making an uproar, thereby can make an uproar to the noise scene of different scenes pertinently, simultaneously, when carrying out the test, the result of test, noise reduction algorithm generates the earphone model of making an uproar that corresponds promptly, thereby can promote the suitability of noise reduction algorithm and every earphone, and can promote the richness of noise simulation data through the mode of constantly enriching scene test type, and then help promoting the user when making an uproar of making an uproar in follow-up use of earphone, the earphone is to the adaptive degree of environment, with the noise reduction effect of promotion earphone.
The third object of the present application is achieved by the following technical solutions:
a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method for testing a noise reducing earpiece described above when the computer program is executed.
The fourth object of the present application is achieved by the following technical solutions:
a computer readable storage medium storing a computer program which, when executed by a processor, performs the steps of the above-described test method of noise reduction headphones.
In summary, the present application includes at least one of the following beneficial technical effects:
1. when the noise reduction algorithm in the noise reduction earphone is tested, corresponding noise simulation data are obtained according to scene test types, noise of different scenes can be simulated, noise reduction tests are conducted, and therefore the noise scenes of the different scenes can be subjected to targeted noise reduction, meanwhile, when the noise reduction algorithm is tested, a corresponding earphone noise reduction model is generated according to test results, namely the noise reduction algorithm, so that the suitability of the noise reduction algorithm and each earphone can be improved, the richness of the noise simulation data can be improved through the mode of continuously enriching the scene test types, and the adaptation degree of the earphone to the environment is improved when the noise reduction earphone is used by a user in a follow-up mode, so that the noise reduction effect of the earphone is improved;
2. because the related noise reduction algorithm consumes larger power consumption when the noise reduction earphone is used for noise reduction, especially the use time of the Bluetooth earphone is influenced, the next scene is tested by the correction result after the noise reduction algorithm model is corrected one by one, and the applicable correction result is used as the noise reduction algorithm of the scene at the same time, so that the number of the noise reduction algorithms in the model of the whole noise reduction algorithm can be reduced, the noise reduction algorithm in the noise reduction process is simplified, and the power consumption of the noise reduction earphone consumed when the noise reduction earphone is used can be reduced;
3. by means of packaging the noise simulation data and associating the noise scene recognition model with the corresponding noise simulation data, the noise scene recognition model and the noise reduction algorithm can be separated, scene recognition calculation can be performed through a terminal connected with the noise reduction earphone through the noise scene recognition model, calculation of the noise reduction earphone in the aspect of noise reduction is further reduced, and therefore use power consumption of the noise reduction earphone is reduced.
Drawings
FIG. 1 is a flow chart of a method of testing a noise reduction earphone according to an embodiment of the present application;
FIG. 2 is a flowchart showing a method step S20 in testing a noise reduction earphone according to an embodiment of the present application;
FIG. 3 is a flowchart showing a method step S23 in testing a noise reduction earphone according to an embodiment of the present application;
FIG. 4 is a flowchart showing a method step S30 in testing a noise reduction earphone according to an embodiment of the present application;
FIG. 5 is a flowchart showing a method step S40 in testing a noise reduction earphone according to an embodiment of the present application;
FIG. 6 is a functional block diagram of a test system for a noise reduction earphone in an embodiment of the present application;
fig. 7 is a schematic view of an apparatus in an embodiment of the present application.
Detailed Description
The present application is described in further detail below with reference to the accompanying drawings.
In an embodiment, as shown in fig. 1, the application discloses a method for testing a noise reduction earphone, which specifically includes the following steps:
s10: and acquiring a scene test type and a noise reduction algorithm model, and acquiring corresponding noise simulation data according to the scene test type.
In the present embodiment, the scene test type refers to a type in which a headset use environment is recorded. Noise simulation data refers to data of simulation noise for each type of scene. The noise reduction algorithm model refers to an initial algorithm that reduces noise.
Specifically, in order to pertinently reduce noise in different scenes during the subsequent use of the noise reduction earphone, an initial algorithm for performing noise reduction processing on noise is acquired first, as a model of the noise reduction algorithm, and by acquiring noise data corresponding to each scene test type as the noise simulation data by acquiring the user on different application scenes using the noise reduction earphone, such as indoor, outdoor, market, various vehicles, and other scenes.
S20: and acquiring audio output reference data, inputting noise simulation data into a noise reduction algorithm model one by one according to the scene test type for testing, and comparing the acquired audio noise reduction result with the audio output reference data respectively to obtain a noise reduction test result.
In this embodiment, the audio output reference data refers to audio data output from the headphones after noise is reduced. The noise reduction test result is the result of the difference between each noise simulation data and the corresponding audio output reference data after the noise reduction algorithm model is used for reducing the noise.
Specifically, from the viewpoint of use of the user, the noise reduction effect of using the noise reduction earphone in each scene is uniform, that is, it is desirable to eliminate noise as much as possible no matter whether the noise reduction earphone is used in any scene, so as to improve the effect of audio playing or be in a quiet state, therefore, after noise reduction is performed on the noise, the data of the audio output meeting the user noise reduction requirement is preset and used as the audio output reference data.
Further, the noise simulation data of each scene test type are input into a noise reduction algorithm model, the noise reduction effect of each scene is obtained through an initial noise reduction algorithm, the noise reduction effect is respectively compared with the audio output reference data, and differences between the noise simulation data and the audio output reference data are used as noise reduction test results after noise in each scene is reduced through a basic noise reduction algorithm of a test primary examination.
S30: and respectively correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type.
Specifically, according to the noise reduction algorithm model, that is, the effect of the initial noise reduction algorithm after the noise reduction is performed on each piece of noise simulation data, the noise reduction algorithm model is respectively corrected, that is, each piece of noise simulation data is respectively corrected in a targeted manner aiming at the noise reduction algorithm model, so that each piece of noise simulation data noise reduction algorithm is obtained, that is, each piece of noise reduction algorithm is corrected by the noise reduction algorithm model according to the noise reduction test result, and the corresponding noise simulation data can be subjected to noise reduction, so that the audio output standard is achieved.
S40: and generating an earphone noise reduction model according to each noise reduction algorithm, and generating model test information according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones.
Specifically, the noise reduction algorithm and the corresponding noise simulation data are packaged, so that a corresponding earphone noise reduction model is obtained, and then when the noise reduction earphone is used for noise reduction, the use scene of a user can be firstly identified through the earphone noise reduction model, namely, the noise environment in which the user is currently located is judged through the acquired noise data of the scene, and the corresponding noise reduction algorithm is further acquired for noise reduction.
Further, according to each type of noise reduction earphone, further test simulation is conducted on the earphone noise reduction model, so that the earphone noise reduction model is better matched with each noise reduction earphone.
In one embodiment, as shown in fig. 2, in step S20, audio output reference data is acquired, noise simulation data is input into a noise reduction algorithm model one by one according to a scene test type for testing, and the acquired audio noise reduction result is compared with the audio output reference data respectively to obtain a noise reduction test result, which specifically includes:
s21: and extracting noise characteristic data corresponding to each piece of noise simulation data, and respectively inputting each piece of noise characteristic data into the noise algorithm model.
In the present embodiment, the noise characteristic data refers to an audio characteristic in each noise analog data.
Specifically, before inputting the noise simulation data to the noise algorithm model, the corresponding noise feature is extracted from the noise simulation data as noise feature data, and the noise feature data is input to the noise algorithm model.
S22: and acquiring an audio noise reduction result corresponding to each noise characteristic data from the noise algorithm model, and extracting each audio noise reduction result to extract the audio noise reduction characteristic data.
Specifically, after the noise characteristic data is input into the noise algorithm model, the noise algorithm model outputs a corresponding audio noise reduction result, and corresponding characteristics are extracted from the noise result to obtain the audio noise reduction characteristic data.
S23: and extracting audio reference features from the audio output reference data, and respectively comparing the audio reference features with each audio noise reduction feature to obtain a noise reduction test result corresponding to each noise feature data.
Specifically, the characteristics of the audio output reference data are extracted and used as the audio reference characteristics, each audio noise reduction characteristic is respectively compared with the audio reference characteristics, and the corresponding comparison result is used as a noise reduction test result.
In one embodiment, as shown in fig. 3, in step S23, namely, extracting audio reference features from audio output reference data, and comparing the audio reference features with each audio noise reduction feature to obtain a noise reduction test result corresponding to each noise feature data, specifically including:
s231: constructing a reference feature vector of the audio reference features and constructing a noise reduction feature vector of each audio noise reduction feature.
Specifically, by means of constructing feature vectors, corresponding feature vectors are constructed for the audio reference feature and each audio noise reduction feature, so that the reference feature vectors and noise reduction feature vectors corresponding to each noise simulation data are obtained.
S232: and calculating comparison distance data of each noise reduction feature vector and the reference feature vector, and taking the comparison distance data as a noise reduction test result.
Specifically, by means of calculating vectors, vector distances between each noise reduction feature vector and the reference feature vector are calculated, comparison distance data are obtained, and therefore differences between each audio noise reduction result and the audio output reference data are judged to be used as noise reduction test results.
In one embodiment, as shown in fig. 4, in step S30, the noise reduction algorithm model is modified according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type, which specifically includes:
s31: and sequencing each scene test type according to a preset rule, and correcting the noise reduction algorithm model with the first sequencing to obtain a first correction result.
Specifically, the scene test types are classified and ordered in advance, and similar types are classified into one type, for example, after the scenes in markets, stations and other rooms are classified into one type, all the scene test types are ordered.
Further, the noise reduction algorithm model corresponding to the first scene test type is corrected, namely, corresponding correction is carried out according to the corresponding noise reduction test result, so that a first correction result is obtained, and after the first noise simulation data is input into the first correction result, the obtained noise reduction result accords with the audio output reference data.
S32: and inputting the noise simulation data corresponding to the second scene test type into the first correction result to obtain correction noise data, and performing difference comparison on the correction noise data and the audio output reference data.
Specifically, the noise simulation data corresponding to the second scene test type is input into the first correction result to obtain corresponding correction noise data, and difference comparison is performed on the correction noise data in the audio output reference data according to the mode of obtaining the noise reduction test result.
S33: and if the difference comparison result is within the preset threshold value, the first correction result is used as a corresponding noise reduction algorithm, otherwise, the second noise reduction algorithm model is corrected to obtain a second correction result.
Specifically, if the difference comparison result is within the preset noise reduction error range, namely within the preset threshold, the first correction result is used as a noise reduction algorithm corresponding to the second scene, otherwise, the noise reduction test result according to the second scene is corrected, and the noise reduction algorithm model is corrected to obtain a second correction result.
S34: starting from the noise simulation data corresponding to the third scene test type, inputting the corresponding noise simulation data into the corrected results after the noise reduction algorithm model is corrected one by one, and comparing the output results with the audio output reference data in a difference mode so as to screen the corresponding corrected results or correct the noise reduction algorithm model.
Specifically, starting from the noise simulation data of the third scene test type, inputting the noise simulation data into the corrected algorithm to test, respectively obtaining corresponding noise reduction results, calculating the difference between each noise reduction result and the audio output reference data, obtaining the noise reduction result with the smallest difference, taking the corresponding correction result as the noise reduction algorithm of the scene if the noise reduction result is within a preset threshold value, and otherwise, carrying out corresponding correction on the noise reduction algorithm model according to the corresponding noise reduction test result.
In one embodiment, as shown in fig. 5, in step S40, a headset noise reduction model is generated according to each noise reduction algorithm, which specifically includes:
s41: and packaging all the noise simulation data to obtain a noise scene recognition model.
Specifically, the noise scene recognition model is obtained by packaging the noise simulation data obtained through statistics, so that the noise of the environment can be obtained in the actual use process of the noise reduction earphone, and the noise of the environment is input into the noise scene recognition model for analysis and recognition, so that the current use environment of the user is recognized, namely, the corresponding scene test type is recognized.
S42: and correlating each noise reduction algorithm with corresponding noise simulation data in the noise scene recognition model to obtain an earphone noise reduction model.
Specifically, according to the association relation between each noise reduction algorithm and the scene test type, the noise reduction algorithm is associated with noise simulation data in the noise scene recognition model, and then the earphone noise reduction model is obtained. In the actual use process, after the user connects the noise reduction earphone with the terminal for audio playing, for example, a smart phone, the noise scene recognition model is sent to the audio playing terminal, the noise of the user using environment is recognized through the audio playing terminal, the recognized result is sent back to the noise reduction earphone, the noise reduction earphone matches the corresponding noise reduction algorithm according to the recognized result to reduce the noise of the environment, and therefore the power consumption of the noise reduction earphone for recognizing the noise environment is saved.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic of each process, and should not limit the implementation process of the embodiment of the present application in any way.
In an embodiment, a test system of a noise reduction earphone is provided, where the test system of the noise reduction earphone corresponds to the test method of the noise reduction earphone in the above embodiment one by one. As shown in fig. 6, the test system of the noise reduction earphone includes a scene simulation module, an algorithm test module, an algorithm correction module and a model test module. The functional modules are described in detail as follows:
the scene simulation module is used for acquiring a scene test type and a noise reduction algorithm model and acquiring corresponding noise simulation data according to the scene test type;
the algorithm test module is used for acquiring audio output reference data, inputting noise simulation data into the noise reduction algorithm model one by one according to the scene test type for testing, and comparing the acquired audio noise reduction result with the audio output reference data respectively to obtain a noise reduction test result;
the algorithm correction module is used for respectively correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type;
the model test module is used for generating an earphone noise reduction model according to each noise reduction algorithm, and generating a model test message according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones.
Optionally, the algorithm test module includes:
the data input sub-module is used for extracting noise characteristic data corresponding to each piece of noise simulation data and respectively inputting each piece of noise characteristic data into the noise algorithm model;
the feature extraction submodule is used for acquiring an audio noise reduction result corresponding to each noise feature data from the noise algorithm model, extracting each audio noise reduction result and extracting the audio noise reduction feature data;
and the feature comparison sub-module is used for extracting audio reference features from the audio output reference data, and respectively comparing the audio reference features with each audio noise reduction feature to obtain a noise reduction test result corresponding to each noise feature data.
Optionally, the feature comparison submodule includes:
the vector construction unit is used for constructing a reference feature vector of the audio reference feature and constructing a noise reduction feature vector of each audio noise reduction feature;
the vector comparison unit is used for calculating comparison distance data of each noise reduction feature vector and the reference feature vector, and taking the comparison distance data as a noise reduction test result.
Optionally, the algorithm correction module includes:
the first correction submodule is used for sequencing each scene test type according to a preset rule, correcting the noise reduction algorithm model of the first sequencing to obtain a first correction result;
the second correction sub-module is used for inputting noise simulation data corresponding to the second scene test type into the first correction result to obtain correction noise data, and performing difference comparison on the correction noise data and the audio output reference data;
the difference comparison sub-module is used for taking the first correction result as a corresponding noise reduction algorithm if the difference comparison result is within a preset threshold value, and otherwise, correcting the second noise reduction algorithm model to obtain a second correction result;
and the cyclic correction sub-module is used for starting from the noise simulation data corresponding to the third scene test type, inputting the corresponding noise simulation data into the corrected results after the noise reduction algorithm model is corrected one by one, and comparing the output results with the audio output reference data in a difference way so as to screen the corresponding corrected results or correct the noise reduction algorithm model.
Optionally, the model test module includes:
the data packaging sub-module is used for packaging all the noise simulation data to obtain a noise scene recognition model;
and the model association sub-module is used for associating each noise reduction algorithm with the corresponding noise simulation data in the noise scene recognition model to obtain the earphone noise reduction model.
For specific limitations regarding the test system of the noise reduction earphone, reference may be made to the above limitations regarding the test method of the noise reduction earphone, and no further description is given here. The modules in the test system of the noise reduction earphone can be realized in whole or in part by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a method of testing a noise reducing earphone.
In one embodiment, a computer device is provided comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of when executing the computer program:
acquiring a scene test type and a noise reduction algorithm model, and acquiring corresponding noise simulation data according to the scene test type;
acquiring audio output reference data, inputting noise simulation data into a noise reduction algorithm model one by one according to a scene test type for testing, and respectively comparing the acquired audio noise reduction result with the audio output reference data to obtain a noise reduction test result;
correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type;
and generating an earphone noise reduction model according to each noise reduction algorithm, and generating model test information according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring a scene test type and a noise reduction algorithm model, and acquiring corresponding noise simulation data according to the scene test type;
acquiring audio output reference data, inputting noise simulation data into a noise reduction algorithm model one by one according to a scene test type for testing, and respectively comparing the acquired audio noise reduction result with the audio output reference data to obtain a noise reduction test result;
correcting the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type;
and generating an earphone noise reduction model according to each noise reduction algorithm, and generating model test information according to the earphone noise reduction model so as to carry out adaptive adjustment according to different noise reduction earphones.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the system is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application.

Claims (6)

Translated fromChinese
1.一种降噪耳机的测试方法,其特征在于,所述降噪耳机的测试方法包括:1. A testing method for noise-canceling headphones, characterized in that the testing method for noise-canceling headphones includes:获取场景测试类型和降噪算法模型,根据所述场景测试类型获取对应的噪声模拟数据;Obtaining a scene test type and a noise reduction algorithm model, and obtaining corresponding noise simulation data according to the scene test type;获取音频输出基准数据,根据所述场景测试类型逐个将所述噪声模拟数据输入至所述降噪算法模型中进行测试,分别将获取到的音频降噪结果与所述音频输出基准数据进行比对,得到降噪测试结果,具体包括:Acquire audio output benchmark data, input the noise simulation data into the noise reduction algorithm model one by one according to the scene test type for testing, and compare the acquired audio noise reduction results with the audio output benchmark data to obtain the noise reduction test results, specifically including:提取每个所述噪声模拟数据对应的噪声特征数据,分别将每个所述噪声特征数据输入至所述降噪算法模型中;Extract the noise characteristic data corresponding to each of the noise simulation data, and input each of the noise characteristic data into the noise reduction algorithm model;从所述降噪算法模型中获取与每个噪声特征数据对应的所述音频降噪结果,并提取每个所述音频降噪结果提取音频降噪特征数据;Obtain the audio noise reduction result corresponding to each noise feature data from the noise reduction algorithm model, and extract the audio noise reduction feature data from each of the audio noise reduction results;从所述音频输出基准数据提取音频基准特征,将所述音频基准特征分别与每个所述音频降噪特征进行特征比对,得到与每种所述噪声特征数据对应的所述降噪测试结果;Extract audio reference features from the audio output reference data, compare the audio reference features with each of the audio noise reduction features, and obtain the noise reduction test results corresponding to each of the noise feature data. ;根据每个所述降噪测试结果分别对所述降噪算法模型进行修正,得到与每个所述场景测试类型对应的噪声降噪算法,具体包括:The noise reduction algorithm model is modified according to each of the noise reduction test results to obtain a noise reduction algorithm corresponding to each of the scene test types, which specifically includes:对每个所述场景测试类型按照预设规则进行排序,对排序第一个的降噪算法模型进行修正,得到第一修正结果;Sort each of the scene test types according to the preset rules, correct the noise reduction algorithm model ranked first, and obtain the first correction result;将第二个的所述场景测试类型对应的噪声模拟数据输入至所述第一修正结果中,得到修正噪声数据,将所述修正噪声数据与音频输出基准数据进行差异比对;Input the noise simulation data corresponding to the second scene test type into the first correction result to obtain corrected noise data, and perform a difference comparison between the corrected noise data and the audio output reference data;若差异比对结果在预设的阈值内,则将第一修正结果作为对应的所述噪声降噪算法,否则对第二个降噪算法模型进行修正,得到第二修正结果;If the difference comparison result is within the preset threshold, the first correction result is used as the corresponding noise reduction algorithm, otherwise the second noise reduction algorithm model is corrected to obtain a second correction result;从第三个所述场景测试类型对应的噪声模拟数据开始,将对应的噪声模拟数据逐个输入至对降噪算法模型进行修正后的修正结果中并将输出的结果与音频输出基准数据进行差异比对,以筛选对应的修正结果或者对降噪算法模型进行修正;Starting from the noise simulation data corresponding to the third scenario test type, the corresponding noise simulation data are input one by one into the correction result after the noise reduction algorithm model is corrected, and the output result is compared with the audio output reference data to screen the corresponding correction result or correct the noise reduction algorithm model;根据每个所述噪声降噪算法生成耳机降噪模型,并根据所述耳机降噪模型生成模型测试消息,以根据不同降噪耳机进行适应性调整,其中,所述根据每个所述噪声降噪算法生成耳机降噪模型,具体包括:Generate a headphone noise reduction model according to each of the noise reduction algorithms, and generate a model test message according to the headphone noise reduction model to perform adaptive adjustment according to different noise reduction headphones, wherein generating a headphone noise reduction model according to each of the noise reduction algorithms specifically includes:将所有噪声模拟数据进行封装,得到噪声场景识别模型;Encapsulate all the noise simulation data to obtain the noise scene recognition model;将每个所述噪声降噪算法与所述噪声场景识别模型中对应的噪声模拟数据进行关联,得到耳机降噪模型。Each of the noise reduction algorithms is associated with the corresponding noise simulation data in the noise scene recognition model to obtain a headphone noise reduction model.2.根据权利要求1所述的降噪耳机的测试方法,其特征在于,所述从所述音频输出基准数据提取音频基准特征,将所述音频基准特征分别与每个所述音频降噪特征进行特征比对,得到与每种所述噪声特征数据对应的所述降噪测试结果,具体包括:2. The testing method of noise reduction headphones according to claim 1, characterized in that: extracting audio reference features from the audio output reference data, and comparing the audio reference features with each of the audio noise reduction features. Conduct feature comparison to obtain the noise reduction test results corresponding to each of the noise characteristic data, specifically including:构建所述音频基准特征的基准特征向量,构建每个所述音频降噪特征的降噪特征向量;Constructing a reference feature vector of the audio reference feature, and constructing a noise reduction feature vector of each of the audio noise reduction features;计算每个所述降噪特征向量与所述基准特征向量的比对距离数据,将所述比对距离数据作为所述降噪测试结果。Comparison distance data between each noise reduction feature vector and the reference feature vector is calculated, and the comparison distance data is used as the noise reduction test result.3.一种降噪耳机的测试系统,其特征在于,所述降噪耳机的测试系统包括:3. A testing system for noise-cancelling headphones, characterized in that the testing system for noise-cancelling headphones includes:场景模拟模块,用于获取场景测试类型和降噪算法模型,根据所述场景测试类型获取对应的噪声模拟数据;A scene simulation module is used to obtain a scene test type and a noise reduction algorithm model, and obtain corresponding noise simulation data according to the scene test type;算法测试模块,用于获取音频输出基准数据,根据所述场景测试类型逐个将所述噪声模拟数据输入至所述降噪算法模型中进行测试,分别将获取到的音频降噪结果与所述音频输出基准数据进行比对,得到降噪测试结果,所述算法测试模块包括:Algorithm test module, used to obtain audio output benchmark data, input the noise simulation data into the noise reduction algorithm model one by one according to the scene test type for testing, and compare the obtained audio noise reduction results with the audio Output the benchmark data for comparison to obtain the noise reduction test results. The algorithm test module includes:数据输入子模块,用于提取每个所述噪声模拟数据对应的噪声特征数据,分别将每个所述噪声特征数据输入至所述降噪算法模型中;A data input submodule is used to extract the noise characteristic data corresponding to each of the noise simulation data, and input each of the noise characteristic data into the noise reduction algorithm model;特征提取子模块,用于从所述降噪算法模型中获取与每个噪声特征数据对应的所述音频降噪结果,并提取每个所述音频降噪结果提取音频降噪特征数据;A feature extraction submodule, configured to obtain the audio noise reduction result corresponding to each noise feature data from the noise reduction algorithm model, and extract the audio noise reduction feature data from each of the audio noise reduction results;特征比对子模块,用于从所述音频输出基准数据提取音频基准特征,将所述音频基准特征分别与每个所述音频降噪特征进行特征比对,得到与每种所述噪声特征数据对应的所述降噪测试结果;Feature comparison sub-module, used to extract audio reference features from the audio output reference data, compare the audio reference features with each of the audio noise reduction features, and obtain the feature data with each of the noise reduction features. The corresponding noise reduction test results;算法校正模块,用于根据每个所述降噪测试结果分别对所述降噪算法模型进行修正,得到与每个所述场景测试类型对应的噪声降噪算法,算法校正模块包括:An algorithm correction module is used to correct the noise reduction algorithm model according to each noise reduction test result to obtain a noise reduction algorithm corresponding to each scene test type. The algorithm correction module includes:第一修正子模块,用于对每个场景测试类型按照预设规则进行排序,对排序第一个的降噪算法模型进行修正,得到第一修正结果;The first correction submodule is used to sort each scene test type according to preset rules, correct the noise reduction algorithm model ranked first, and obtain the first correction result;第二修正子模块,用于将第二个的场景测试类型对应的噪声模拟数据输入至第一修正结果中,得到修正噪声数据,将修正噪声数据与音频输出基准数据进行差异比对;A second correction submodule, used for inputting the noise simulation data corresponding to the second scene test type into the first correction result to obtain corrected noise data, and performing a difference comparison between the corrected noise data and the audio output reference data;差异比对子模块,用于若差异比对结果在预设的阈值内,则将第一修正结果作为对应的噪声降噪算法,否则对第二个降噪算法模型进行修正,得到第二修正结果;The difference comparison sub-module is used to use the first correction result as the corresponding noise reduction algorithm if the difference comparison result is within the preset threshold. Otherwise, the second noise reduction algorithm model is corrected to obtain the second correction. result;循环修正子模块,用于从第三个场景测试类型对应的噪声模拟数据开始,将对应的噪声模拟数据逐个输入至对降噪算法模型进行修正后的修正结果中并将输出的结果与音频输出基准数据进行差异比对,以筛选对应的修正结果或者对降噪算法模型进行修正;The loop correction submodule is used to start from the noise simulation data corresponding to the third scene test type, input the corresponding noise simulation data one by one into the correction results after correcting the noise reduction algorithm model, and combine the output results with the audio output Compare the differences between the benchmark data to screen the corresponding correction results or correct the noise reduction algorithm model;模型测试模块,用于根据每个所述噪声降噪算法生成耳机降噪模型,并根据所述耳机降噪模型生成模型测试消息,以根据不同降噪耳机进行适应性调整,模型测试模块包括:A model testing module is used to generate a headphone noise reduction model according to each of the noise reduction algorithms, and to generate a model testing message according to the headphone noise reduction model, so as to perform adaptive adjustments according to different noise reduction headphones. The model testing module includes:数据封装子模块,用于将所有噪声模拟数据进行封装,得到噪声场景识别模型;The data encapsulation sub-module is used to encapsulate all noise simulation data to obtain the noise scene recognition model;模型关联子模块,用于将每个噪声降噪算法与噪声场景识别模型中对应的噪声模拟数据进行关联,得到耳机降噪模型。The model association submodule is used to associate each noise reduction algorithm with the corresponding noise simulation data in the noise scene recognition model to obtain the headphone noise reduction model.4.根据权利要求3所述的降噪耳机的测试系统,其特征在于,所述特征比对子模块包括:4. The noise reduction headphone testing system according to claim 3, wherein the feature comparison submodule comprises:向量构建单元,用于构建所述音频基准特征的基准特征向量,构建每个所述音频降噪特征的降噪特征向量;A vector construction unit, configured to construct a reference feature vector of the audio reference feature and a noise reduction feature vector of each of the audio noise reduction features;向量比对单元,用于计算每个所述降噪特征向量与所述基准特征向量的比对距离数据,将所述比对距离数据作为所述降噪测试结果。A vector comparison unit, configured to calculate comparison distance data between each of the noise reduction feature vectors and the reference feature vector, and use the comparison distance data as the noise reduction test result.5.一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序,其特征在于,所述处理器执行所述计算机程序时实现如权利要求1至2任一项所述降噪耳机的测试方法的步骤。5. A computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that when the processor executes the computer program, the processor implements the claims as claimed in Steps of the noise-cancelling headphone testing method described in any one of 1 to 2.6.一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其特征在于,所述计算机程序被处理器执行时实现如权利要求1至2任一项所述降噪耳机的测试方法的步骤。6. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the noise-cancelling headphone testing method according to any one of claims 1 to 2.
CN202311232265.XA2023-09-212023-09-21 A testing method and system for noise reduction headphonesActiveCN117202071B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202311232265.XACN117202071B (en)2023-09-212023-09-21 A testing method and system for noise reduction headphones

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202311232265.XACN117202071B (en)2023-09-212023-09-21 A testing method and system for noise reduction headphones

Publications (2)

Publication NumberPublication Date
CN117202071A CN117202071A (en)2023-12-08
CN117202071Btrue CN117202071B (en)2024-03-29

Family

ID=88983339

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202311232265.XAActiveCN117202071B (en)2023-09-212023-09-21 A testing method and system for noise reduction headphones

Country Status (1)

CountryLink
CN (1)CN117202071B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
WO2019045304A1 (en)*2017-08-312019-03-07주식회사 글로베인Anc test module and anc test device using same
CN111800694A (en)*2020-06-302020-10-20深圳市豪恩声学股份有限公司Filter design method and device of active noise reduction earphone and test equipment
CN113115195A (en)*2021-04-222021-07-13东莞市声强电子有限公司Automatic production test method and device for active noise reduction earphone
CN113327626A (en)*2021-06-232021-08-31深圳市北科瑞声科技股份有限公司Voice noise reduction method, device, equipment and storage medium
CN114205724A (en)*2021-12-092022-03-18科大讯飞股份有限公司Debugging method, device and equipment of hearing aid earphone

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
WO2019045304A1 (en)*2017-08-312019-03-07주식회사 글로베인Anc test module and anc test device using same
CN111800694A (en)*2020-06-302020-10-20深圳市豪恩声学股份有限公司Filter design method and device of active noise reduction earphone and test equipment
CN113115195A (en)*2021-04-222021-07-13东莞市声强电子有限公司Automatic production test method and device for active noise reduction earphone
CN113327626A (en)*2021-06-232021-08-31深圳市北科瑞声科技股份有限公司Voice noise reduction method, device, equipment and storage medium
CN114205724A (en)*2021-12-092022-03-18科大讯飞股份有限公司Debugging method, device and equipment of hearing aid earphone

Also Published As

Publication numberPublication date
CN117202071A (en)2023-12-08

Similar Documents

PublicationPublication DateTitle
US11189287B2 (en)Optimization method, apparatus, device for wake-up model, and storage medium
US10332507B2 (en)Method and device for waking up via speech based on artificial intelligence
CN109862393B (en) Soundtrack method, system, device and storage medium for video file
WO2020014899A1 (en)Voice control method, central control device, and storage medium
CN108831481A (en)Symbol adding method, device, computer equipment and storage medium in speech recognition
CN112163479B (en) Motion detection method, device, computer equipment and computer readable storage medium
CN109522550B (en)Text information error correction method and device, computer equipment and storage medium
WO2020125397A1 (en)Audio data pushing method and device, computer apparatus, and storage medium
WO2020125386A1 (en)Expression recognition method and apparatus, computer device, and storage medium
CN109460891A (en)Data processing method, device and computer equipment based on satisfaction evaluation
CN109074804B (en)Accent-based speech recognition processing method, electronic device, and storage medium
CN109101216A (en)Audio method of adjustment, device, electronic equipment and storage medium
CN113782004B (en)Man-machine voice interaction testing method and device, computer equipment and storage medium
US9929709B1 (en)Electronic device capable of adjusting output sound and method of adjusting output sound
CN112687266B (en)Speech recognition method, device, computer equipment and storage medium
CN110581918A (en)Voice equipment, control method thereof, server and computer storage medium
CN110851477B (en)Stream data processing method, stream data processing device, computer equipment and storage medium
CN110163151B (en)Training method and device of face model, computer equipment and storage medium
CN114743067A (en) Training data enhancement method, device, computer equipment and storage medium
CN109683915B (en)Parameter modification method and device, computer equipment and storage medium
CN112634932B (en)Audio signal processing method and device, server and related equipment
CN114493902A (en)Multi-mode information anomaly monitoring method and device, computer equipment and storage medium
CN115881103A (en)Voice emotion recognition model training method, voice emotion recognition method and device
CN112395857B (en)Speech text processing method, device, equipment and medium based on dialogue system
CN117202071B (en) A testing method and system for noise reduction headphones

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant
PE01Entry into force of the registration of the contract for pledge of patent right
PE01Entry into force of the registration of the contract for pledge of patent right

Denomination of invention:A testing method and system for noise cancelling headphones

Granted publication date:20240329

Pledgee:China Co. truction Bank Corp Guangzhou green finance reform and innovation pilot area Huadu Branch

Pledgor:Guangdong jinhaina Industrial Co.,Ltd.

Registration number:Y2025980022060


[8]ページ先頭

©2009-2025 Movatter.jp