A testing method and system for howling detection function

Abstract

The present invention discloses a testing method and system for a howling detection function. A terminal management module determines application information to be tested and a test audio file list from a background management module according to a test task instruction, downloads the test audio file from the background management module according to the test audio file list, starts the application to be tested according to the application information to be tested, and uses the test audio file to test the howling detection function of the application to be tested to obtain a test result. In this way, there is no need for manual shouting simulation or manual operation test. The application to be tested can be automatically started and the howling detection function can be tested using the test audio file, thereby realizing automated testing of the howling detection function and ensuring the reliability of the test result, thereby more efficiently testing the howling detection function.

Description

Technical Field

[0001] The present invention relates to the technical field of audio detection, and in particular to a testing method and system for a howling detection function. Background Art

[0002] With the rapid development of the internet and technological advancements, audio and video have achieved unprecedented widespread application, especially in scenarios such as online conferencing, voice interaction in games, and live streaming. Among them, the critical audio problem of howling is a pain point that plagues users and a challenge for audio engineers. Howling is caused by the self-excited amplification of sound energy, which can greatly affect the user experience and even cause noise pollution. Current audio interactive applications all have howling issues, especially when multiple terminal devices are in close proximity. Currently, most manufacturers use howling detection prompts to guide users to turn off the microphone / speaker to avoid continuous howling. Of course, some manufacturers also use howling suppression methods to reduce the severity of howling, but howling detection remains an indispensable step.

[0003] However, the howling detection function may also have side effects. For example, a sound that is not actually a howling sound may be falsely detected by the howling detection program, affecting the user experience. Therefore, the premise for releasing the howling detection function is to ensure that it has a sufficiently high true detection rate and a sufficiently low false detection rate. Therefore, dedicated evaluation tests are required. Often, R&D personnel need to repeatedly verify whether various sounds trigger howling detection. Currently, the main verification method is for testers to shout into the microphone, simulate or play the corresponding voice through pre-set audio scenarios, and then unify the results.

[0004] This method of testing the howling detection function has the following problems:

[0005] 1. The most common triggering scenario for howling detection is when two devices are close together and a human voice is heard. After generating a howling sound, the sound needs to be collected and detected as howling. This process mainly relies on manual microphone simulation, which can easily affect other colleagues in a multi-person office. Furthermore, even if the devices are placed in the same position each time, it is difficult to trigger the same howling effect due to the influence of the external environment, and the consistency of the test data cannot be guaranteed.

[0006] 2. The conventional false detection rate refers to the percentage of sounds that are falsely detected as howling. This type of test data can be very extensive, and many of them are not human voices, but may be falsely detected. Examples include background music, the sounds of doors opening and closing, whistles, 110 sirens, the sounds of walking, the noisy sounds of people discussing, and so on. The number of types can reach thousands. After testers collect so much audio, playing and observing it takes a considerable amount of time.

[0007] 3. Howling detection often involves artificial intelligence algorithms, which are inherently not 100% reliable. The standard for evaluating its suitability for release is often compared with competing products. Therefore, manually verifying the true / false detection rates of competing products undoubtedly increases the workload.

[0008] 4. When the audio data used in the test is very long and a sufficient number of tests are required, hundreds of hours of data may be required. Manual testing is inefficient. Furthermore, even if a specific audio file is known to trigger howling, it is difficult to quickly locate the audio clip that causes the howling. Summary of the Invention

[0009] The technical problem to be solved by the present invention is to provide a method and system for testing a howling detection function, which can test the howling detection function more efficiently.

[0010] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0011] A method for testing a howling detection function comprises the steps of:

[0012] The terminal management module receives the test task instructions issued by the background management module;

[0013] The terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list;

[0014] The terminal management module starts the application to be tested according to the application to be tested information, and uses the test audio file to test the howling detection function of the application to be tested to obtain a test result.

[0015] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0016] A testing system for a howling detection function includes a background management module and a terminal management module. The background management module includes a first memory, a first processor, and a first computer program stored in the first memory and executable on the first processor. The terminal management module includes a second memory, a second processor, and a second computer program stored in the second memory and executable on the second processor. When the first processor executes the first computer program, the steps performed by the background management module in the above-mentioned method for testing a howling detection function are implemented.

[0017] When the second processor executes the second computer program, the steps performed by the terminal management module in the above-mentioned method for testing the howling detection function are implemented.

[0018] The beneficial effects of the present invention are: the terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list, starts the application to be tested according to the application information to be tested, and uses the test audio file to test the howling detection function of the application to be tested to obtain the test result. In this way, there is no need for manual shouting simulation or manual operation test. The application to be tested can be automatically started and the howling detection function can be tested using the test audio file, thereby realizing the automated testing of the howling detection function and ensuring the reliability of the test results, thereby more efficiently testing the howling detection function. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flowchart of a method for testing a howling detection function according to an embodiment of the present invention;

[0020] Figure 2 FIG. 4 is a structural diagram of a test system for a howling detection function according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0022] Please refer to Figure 1 , a method for testing a howling detection function, comprising the steps of:

[0023] The terminal management module receives the test task instructions issued by the background management module;

[0024] The terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list;

[0025] The terminal management module starts the application to be tested according to the application to be tested information, and uses the test audio file to test the howling detection function of the application to be tested to obtain a test result.

[0026] From the above description, it can be seen that the beneficial effect of the present invention is that: the terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list, opens the application to be tested according to the application information to be tested, and uses the test audio file to test the howling detection function of the application to be tested to obtain the test results. In this way, there is no need for manual shouting simulation or manual operation test. The application to be tested can be automatically opened and the howling detection function can be tested using the test audio file, thereby realizing the automated testing of the howling detection function and ensuring the reliability of the test results, thereby more efficiently testing the howling detection function.

[0027] Furthermore, before the terminal management module receives the test task issued by the background management module, it also includes:

[0028] The backend management module determines the test type and selects a test audio file set from a preset database;

[0029] The background management module determines the application to be tested of the terminal to be tested, and allocates the test audio file set according to the application to be tested;

[0030] The background management module generates a test task according to the test type, the test audio file set and the application to be tested;

[0031] The background management module determines the task ID and the test range ID of the test task, generates a test task instruction according to the task ID and the test range ID, and sends the test task instruction to the terminal management module corresponding to the terminal to be tested.

[0032] From the above description, it can be seen that the background management module first automatically generates tasks according to user needs, and sends the test tasks to the terminal management module, so as to subsequently realize the automated testing of the howling detection function. The background management module can also determine the applications to be tested for multiple terminals to be tested, so as to test the howling detection function of the applications of multiple terminals in one test process, thereby improving the test efficiency.

[0033] Furthermore, the terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, including:

[0034] The terminal management module sends the task ID and the test range ID to the background management module;

[0035] The background management module determines the application information to be tested according to the task ID, and determines the test audio file list according to the test range ID and the test audio file set;

[0036] The background management module returns the application information to be tested and the test audio file list to the terminal management module.

[0037] As can be seen from the above description, the background management module returns the application information to be tested and the test audio file list to the terminal management module so that the terminal management module can determine the application to be tested and the test audio file to be downloaded, thereby realizing automated testing.

[0038] Furthermore, downloading the test audio file from the background management module according to the test audio file list includes:

[0039] The terminal management module queries whether a corresponding test audio file already exists locally according to the test audio file list. If so, the existing audio file is not downloaded. If not, the test audio file is downloaded from the background management module.

[0040] As can be seen from the above description, when the corresponding test audio file already exists locally, it will not be downloaded again, thereby avoiding repeated downloading of files and improving the test efficiency of the howling detection function.

[0041] Furthermore, the information of the application to be tested includes a download address of the test application and a download address of an automated activation program of the test application;

[0042] The terminal management module starts the application to be tested according to the application to be tested information, including:

[0043] The terminal management module determines whether there is a download record locally based on the test application download address and the test application's automated activation program download address; if so, the automated activation program is used to start the application to be tested; if not, the application to be tested and the automated activation program are downloaded based on the test application download address and the test application's automated activation program download address, and the automated startup program is used to start the application to be tested.

[0044] From the above description, it can be seen that when there are download records of the application to be tested and the automation activation program locally, the download is not performed, which improves the test efficiency. Otherwise, the application to be tested and the automation activation program are downloaded according to the test application download address and the test application's automation activation program download address for subsequent testing.

[0045] Furthermore, the test result includes the task ID, the test audio file ID, the test time and the test result of the howling detection function;

[0046] After testing the howling detection function of the application to be tested using the test audio file and obtaining the test result, the method further includes:

[0047] The terminal management module sends the task ID, the test audio file ID, the test time, and the detection result of the howling detection function to the background management module;

[0048] The background management module collects statistics on all received test results according to the task ID to obtain a detection rate of the howling detection function.

[0049] From the above description, it can be seen that the terminal management module sends the task ID, test audio file ID, test time and detection results of the howling detection function to the background management module. The background management module counts all the test results received according to the task ID to obtain the detection rate of the howling detection function, which can intuitively display the test results and ensure the effectiveness of the test.

[0050] Furthermore, after the background management module collects statistics on all received test results according to the task ID and obtains the detection rate of the howling detection function, it further includes:

[0051] The background management module receives a detection rate comparison request, where the detection rate comparison request includes at least two test applications to be compared;

[0052] The background management module determines the difference test audio files of the at least two to-be-compared test applications, and displays the detection rates of the howling detection functions of the at least two to-be-compared test applications and the difference test audio files.

[0053] From the above description, it can be seen that when it is necessary to compare the detection rates of different test applications, the detection rates of the howling detection functions of at least two test applications to be compared and the difference test audio files can be displayed so that users can intuitively understand the defects of the howling detection functions of different test applications.

[0054] Furthermore, before the terminal management module receives the test task issued by the background management module, it also includes:

[0055] The backend management module receives the audio file uploaded by the user and determines whether the audio file exceeds a preset duration. If so, the audio file is sliced ​​to obtain a test audio file. If not, the audio file is used as a test audio file.

[0056] The background management module classifies the test audio files and adds the classified test audio files to a preset database.

[0057] From the above description, it can be seen that when the audio file uploaded by the user exceeds the preset length, the audio file is sliced ​​to obtain a test audio file. In subsequent tests, even if the long audio file is used for testing, the audio segment that triggers the howling can be quickly located.

[0058] Furthermore, it also includes:

[0059] The terminal management module determines whether the application to be tested crashes or the microphone activation fails, and if so, generates abnormal feedback information and sends the abnormal feedback information to the background management module;

[0060] The background management module notifies the user of the abnormal feedback module according to a preset notification method.

[0061] As can be seen from the above description, if anomalies such as program crash or microphone activation failure occur during the test, the abnormal feedback information can be reported in time so that the background management module can notify the user, ensuring the effectiveness of the test.

[0062] Please refer to Figure 2 A test system for a howling detection function includes a background management module and a terminal management module. The background management module includes a first memory, a first processor, and a first computer program stored in the first memory and executable on the first processor. The terminal management module includes a second memory, a second processor, and a second computer program stored in the second memory and executable on the second processor. When the first processor executes the first computer program, the steps performed by the background management module in the above-mentioned method for testing a howling detection function are implemented.

[0063] When the second processor executes the second computer program, the steps performed by the terminal management module in the above-mentioned method for testing the howling detection function are implemented.

[0064] The above-mentioned method and system for testing the howling detection function of the present invention can be applied to scenarios where the howling detection function needs to be tested. The following is an explanation through specific implementation methods:

[0065] Please refer to Figure 1 , embodiment 1 of the present invention is:

[0066] A method for testing a howling detection function comprises the steps of:

[0067] S1. The background management module receives the audio file uploaded by the user and determines whether the audio file exceeds the preset duration. If so, the audio file is sliced ​​to obtain a test audio file. If not, the audio file is used as a test audio file.

[0068] The audio files may be one or more, and may include howling audio, human voice noise audio, and / or other audio, etc. The preset duration is flexibly set according to actual conditions. In an optional implementation, the preset duration is 10 seconds.

[0069] For example, a 5-minute audio file is split into 5 × 60 ÷ 10 = 30 segments by default, each identified by the original file name + an incrementing subscript value, such as audio_1, audio_2, and so on. Of course, if the audio is too fragmented, for example, if the last segment is 15 seconds and there are no further 10-second segments to split into, no further segments will be created. This fragmentation of test audio data facilitates batch testing and allows for quick location of the audio segment that triggers howling.

[0070] S2. The background management module classifies the test audio files and adds the classified test audio files to a preset database.

[0071] S3. The background management module determines the test type and selects a test audio file set from a preset database.

[0072] The test type includes a true detection or a false detection, wherein the true detection is the detection rate of samples that need to be detected in the test, and the false detection is the detection rate of samples that do not need to be detected in the test.

[0073] S4. The background management module determines the application to be tested of the terminal to be tested, and allocates the test audio file set according to the application to be tested.

[0074] The terminals to be tested can be multiple or single. When there are multiple terminals to be tested, the applications to be tested of multiple terminals to be tested can be tested simultaneously, which greatly improves the test efficiency. The test audio file set includes two types of audio: howling audio and non-howling audio.

[0075] When there are multiple terminals to be tested, the test audio file set is allocated one by one to the application to be tested of the terminal to be tested, and the test audio files are different from each other. The application to be tested is an application with a howling detection function.

[0076] Specifically, the background management module determines the application to be tested and its version of the terminal to be tested.

[0077] In an optional implementation, when the background management module determines the application to be tested of the terminal to be tested, it further includes: whether to restart the test application information when the configuration is detected, and the default is no.

[0078] S5. The background management module generates a test task according to the test type, the test audio file set, and the application to be tested.

[0079] S6. The background management module determines the task ID and the test range ID of the test task, generates a test task instruction according to the task ID and the test range ID, and sends the test task instruction to the terminal management module corresponding to the terminal to be tested.

[0080] For example, when there is only one terminal to be tested, the test range ID is 1. When there are multiple terminals to be tested, the task IDs are the same, but the test range IDs increase one by one. Assuming there are 5 terminals to be tested, the test range IDs are 1, 2, 3, 4, and 5 respectively.

[0081] S7. The terminal management module receives the test task instruction issued by the background management module.

[0082] In an optional embodiment, after the terminal management module is started, it sends a login request including MAC address, IP, etc. to the background management module; the background management module determines whether the MAC address is a new device, and if so, registers a new login number and returns the device login number and login address; if not, obtains the login number and login address corresponding to the MAC address, and returns the device login number and login address; the terminal management module carries the login number and initiates a Socket long link to the login address to realize two-way instant communication between the background management module and the terminal management module. When the Socket long link between the background management module and the terminal management module is online, it is marked as online, and when the link is disconnected, it is marked as offline.

[0083] S8, the terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list, specifically including S81-S83:

[0084] S81. The terminal management module sends the task ID and the test range ID to the background management module.

[0085] S82. The background management module determines the application information to be tested according to the task ID, and determines a test audio file list according to the test range ID and the test audio file set.

[0086] The information of the application to be tested includes a download address of the test application and a download address of an automatic activation program of the test application.

[0087] When the test range ID is 1, it means that the terminal to be tested is single. At this time, the test audio file list is directly determined according to the test audio file set. When the test range ID is multiple, the test audio file range allocated to the last terminal to be tested is Files[Limit*(Client_Count-1),File_Count], and the test audio file range allocated to other terminals to be tested is Files[Limit*(Client_Count-1),Limit*Client_Count], where File_Count represents the number of files in the test audio file set, and Limit represents the offset, Limit=File_Count / / Client_Count, and Client_Count represents the number of terminals to be tested. For example, the test audio file set includes 999 audio files, and the maximum test range ID is 5. Then the test audio file serial numbers allocated to these 5 terminals to be tested are 1-200, 201-400, 401-600, 601-800, and 801-999, respectively, which are used as the test audio file list.

[0088] S83. The background management module returns the application information to be tested and the test audio file list to the terminal management module.

[0089] S84. The terminal management module queries whether a corresponding test audio file already exists locally according to the test audio file list. If so, the existing audio file is not downloaded. If not, the test audio file is downloaded from the background management module.

[0090] S9, the terminal management module starts the application to be tested according to the application to be tested information, and uses the test audio file to test the howling detection function of the application to be tested to obtain a test result, specifically including S91-S92:

[0091] S91. The terminal management module determines whether there is a download record locally based on the test application download address and the test application's automated activation program download address. If so, the automated activation program is used to start the application to be tested. If not, the application to be tested and the automated activation program are downloaded based on the test application download address and the test application's automated activation program download address, and the automated startup program is used to start the application to be tested.

[0092] The automatic startup procedure includes selecting High Definition Audio Device as the microphone acquisition procedure - turning on the microphone - setting the system's speakers to High Definition Audio Device.

[0093] S92: The terminal management module uses the test audio file to test the howling detection function of the application to be tested to obtain a test result.

[0094] The test result includes the task ID, the test audio file ID, the test time and the test result of the howling detection function.

[0095] Specifically, the terminal management module calls pyaudio (audio processing toolkit) to play the test audio files in sequence to test the howling detection function of the application to be tested, and records whether the howling detection pop-up window is triggered after each test audio file is played. If so, the result of the local audio file playback is recorded as 1 (normalization identifier, which is also convenient for the background to count the total number of occurrences), otherwise it is 0 to obtain the test result.

[0096] In an optional embodiment, when the howling detection pop-up window is triggered, it is also determined whether the information "whether to restart the test application when detected" is yes. If so, the application to be tested is restarted. If not, the howling detection pop-up window is closed and the microphone is restarted. This is because some howling detection algorithms are causal, that is, after triggering a howling detection, it may make subsequent howling detections more sensitive, so it is necessary to restart the application for evaluation. However, some algorithms will not affect subsequent tests, so there is no need to restart the application. For a more comprehensive evaluation, both modes will be considered for verification.

[0097] In an optional embodiment, the method further includes:

[0098] The terminal management module determines whether the application to be tested crashes or the microphone activation fails, and if so, generates abnormal feedback information and sends the abnormal feedback information to the background management module;

[0099] The background management module notifies the user of the abnormal feedback module according to a preset notification method.

[0100] In an optional implementation, the preset notification method is email.

[0101] S10. The terminal management module sends the task ID, the test audio file ID, the test time, and the detection result of the howling detection function to the background management module.

[0102] S11 . The background management module collects statistics on all received test results according to the task ID to obtain a detection rate of the howling detection function.

[0103] S12. The background management module receives a detection rate comparison request, where the detection rate comparison request includes at least two test applications to be compared.

[0104] The two applications to be compared and tested are different versions of the same application or different applications.

[0105] S13. The background management module determines the difference test audio files of the at least two to-be-compared test applications, and displays the detection rates of the howling detection functions of the at least two to-be-compared test applications and the difference test audio files.

[0106] In practical applications, testers only need to record or collect various howling audio files in advance and save them to a preset database. By automatically playing these audio files on a computer and recording the results during each test round, testers no longer need to manually trigger howling detection by holding multiple devices at their workstations. This ensures that the corresponding audio howling detection verification is completed without any noise, while also ensuring the consistency of input test data for each round. Furthermore, testers can select multiple terminals for testing simultaneously in the background. If the original test time for 10 terminals is 10 hours, selecting them simultaneously can shorten the test time to about one hour. The test results from all 10 terminals are returned and compiled into a single test report, allowing for rapid evaluation of a large number of audio files. When a sufficient number of test devices are available, this can be achieved within a sufficiently short delivery cycle. Furthermore, test comparison is supported. Testers simply upload the third-party application program and startup script, execute the test, and obtain test results for competing third-party products. The tester can then compare the test results of their company's product with those of the competing third-party product, using this as a reference standard for development and optimization. At the same time, when playing audio, since audio files exceeding the preset length can be sliced, it is possible to accurately locate where the problem occurs in the audio file. By directly clicking on the corresponding audio file to play and listen, you will know what type of audio triggered the howling.

[0107] Please refer to Figure 2 , the second embodiment of the present invention is:

[0108] A testing system for a howling detection function includes a background management module and a terminal management module. The background management module includes a first memory, a first processor, and a first computer program stored in the first memory and executable on the first processor. The terminal management module includes a second memory, a second processor, and a second computer program stored in the second memory and executable on the second processor. When the first processor executes the first computer program, the steps performed by the background management module in the above-mentioned method for testing a howling detection function are implemented.

[0109] When the second processor executes the second computer program, the steps performed by the terminal management module in the above-mentioned method for testing the howling detection function are implemented.

[0110] In summary, the present invention provides a method and system for testing a howling detection function. The terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list. The application to be tested is started according to the application information to be tested, and the test audio file is used to test the howling detection function of the application to be tested to obtain the test result. In this way, there is no need for manual shouting simulation or manual operation test. The application to be tested can be automatically started and the test audio file can be used to test the howling detection function, thereby realizing howling detection. The automated testing of the function ensures the reliability of the test results, thereby testing the howling detection function more efficiently. In addition, when it is necessary to compare the detection rates of different test applications, the detection rates of the howling detection functions of at least two test applications to be compared and the difference test audio files can be displayed, so that users can intuitively understand the defects of the howling detection functions of different test applications. Moreover, when the audio file uploaded by the user exceeds the preset length, the audio file is sliced ​​to obtain a test audio file. In subsequent tests, even if the long audio file is used for testing, the audio segment that triggers the howling can be quickly located.

[0111] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for testing a howling detection function, characterized in that: Including steps: The backend management module determines the test type and selects a test audio file set from a preset database; The background management module determines the application to be tested of the terminal to be tested, and allocates the test audio file set according to the application to be tested; The background management module generates a test task according to the test type, the test audio file set and the application to be tested; The background management module determines the task ID and the test range ID of the test task, generates a test task instruction according to the task ID and the test range ID, and sends the test task instruction to the terminal management module corresponding to the terminal to be tested; The terminal management module receives the test task instructions issued by the background management module; The terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, and downloads the test audio file from the background management module according to the test audio file list; The terminal management module starts the application to be tested according to the application to be tested information, and uses the test audio file to test the howling detection function of the application to be tested to obtain a test result.

2. The method for testing a howling detection function according to claim 1, wherein: The terminal management module determines the application information to be tested and the test audio file list from the background management module according to the test task instruction, including: The terminal management module sends the task ID and the test range ID to the background management module; The background management module determines the application information to be tested according to the task ID, and determines the test audio file list according to the test range ID and the test audio file set; The background management module returns the application information to be tested and the test audio file list to the terminal management module.

3. The method for testing a howling detection function according to claim 1, wherein: The downloading of the test audio file from the background management module according to the test audio file list includes: The terminal management module queries whether a corresponding test audio file already exists locally according to the test audio file list. If so, the existing audio file is not downloaded. If not, the test audio file is downloaded from the background management module.

4. The method for testing a howling detection function according to claim 1, wherein: The application information to be tested includes a download address of the test application and a download address of an automated activation program of the test application; The terminal management module starts the application to be tested according to the application to be tested information, including: The terminal management module determines whether there is a download record locally based on the test application download address and the test application's automated activation program download address; if so, the automated activation program is used to activate the application to be tested; if not, the application to be tested and the automated activation program are downloaded based on the test application download address and the test application's automated activation program download address, and the automated activation program is used to activate the application to be tested.

5. The method for testing a howling detection function according to claim 1, wherein: The test results include the task ID, the test audio file ID, the test time, and the test results of the howling detection function; After testing the howling detection function of the application to be tested using the test audio file and obtaining the test result, the method further includes: The terminal management module sends the task ID, the test audio file ID, the test time, and the detection result of the howling detection function to the background management module; The background management module collects statistics on all received test results according to the task ID to obtain a detection rate of the howling detection function.

6. The method for testing a howling detection function according to claim 5, wherein: The background management module collects statistics on all received test results according to the task ID to obtain a detection rate of the howling detection function, and further includes: The background management module receives a detection rate comparison request, where the detection rate comparison request includes at least two test applications to be compared; The background management module determines the difference test audio files of the at least two to-be-compared test applications, and displays the detection rates of the howling detection functions of the at least two to-be-compared test applications and the difference test audio files.

7. The method for testing a howling detection function according to claim 1, wherein: Before the terminal management module receives the test task issued by the background management module, the terminal management module further includes: The backend management module receives the audio file uploaded by the user and determines whether the audio file exceeds a preset duration. If so, the audio file is sliced ​​to obtain a test audio file. If not, the audio file is used as a test audio file. The background management module classifies the test audio files and adds the classified test audio files to a preset database.

8. The method for testing a howling detection function according to claim 1, wherein: Also includes: The terminal management module determines whether the application to be tested crashes or the microphone activation fails, and if so, generates abnormal feedback information and sends the abnormal feedback information to the background management module; The background management module notifies the user of the abnormal feedback information according to a preset notification method.

9. A test system for a howling detection function, comprising a background management module and a terminal management module, wherein the background management module comprises a first memory, a first processor, and a first computer program stored in the first memory and executable on the first processor, and the terminal management module comprises a second memory, a second processor, and a second computer program stored in the second memory and executable on the second processor, characterized in that: When the first processor executes the first computer program, the steps performed by the background management module in the method for testing a howling detection function according to any one of claims 1 to 8 are implemented; When the second processor executes the second computer program, the steps performed by the terminal management module in the method for testing a howling detection function according to any one of claims 1 to 8 are implemented.

Images