Reverberation testing method, system, computer device and storage medium

Abstract

The embodiment of the present application discloses a reverberation test method, system, computer device and storage medium; the embodiment of the present application includes: through the reverberation module, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated; through the processing module, the reverberation signal is processed to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets the preset similarity; through the test module, a test audio library is established according to the target reverberation signal, and the test audio library includes test audio corresponding to the target reverberation signal; through the test module, the target test audio is determined from the test audio library; through the test module, the target test audio is played to test the target device. Therefore, this solution can generate test audio for different reverberation scenarios, can be widely used in reverberation test scenarios, and improve the efficiency of reverberation testing.

Description

Technical Field

[0001] The present application relates to the field of acoustic technology, and in particular to a reverberation testing method, system, computer device and storage medium. Background Art

[0002] Reverberation, also known as reverberation, refers to the mixing phenomenon caused by the continuation of sound after the sound source stops making sound. Reverberation is often formed by the continuous reflection of sound into multiple echoes, which are mixed together. The duration is usually measured by the time required for the average sound energy density to decrease to one millionth of the original value, which is called reverberation time.

[0003] In recent years, with the development of intelligent voice recognition products such as smart speakers and smart TVs, more and more smart products recognize user voices by setting up microphones. However, using microphones to recognize voice signals indoors will inevitably be interfered by reflected signals from indoor walls, ceilings and other obstacles, so that the voice signals will be linearly distorted. This distortion is usually called reverberation, which will degrade the fidelity and intelligibility of the voice, causing the performance of the voice communication system and the automatic voice recognition system to decline. It can be seen that in the research and development process of intelligent voice recognition products, a large number of reverberation voice scene tests are required to ensure that the product can meet the use requirements of the reverberation scene.

[0004] Therefore, how to efficiently conduct reverberation testing on intelligent speech recognition products is a major issue that needs to be solved at present. Summary of the invention

[0005] The embodiments of the present application provide a reverberation test method, system, computer device, and storage medium, which can generate test audio for different reverberation scenarios, can be widely used in reverberation test scenarios, and improve the efficiency of reverberation testing.

[0006] An embodiment of the present application provides a reverberation test method, which is applied to a reverberation test system. The reverberation test system includes a reverberation module, a processing module and a testing module. The method includes: obtaining reverberation audio according to original sound audio simulation by the reverberation module, and generating a reverberation signal and an original sound signal; processing the reverberation signal by the processing module to obtain a target reverberation signal, wherein the similarity between the target reverberation signal and the original sound signal satisfies a preset similarity; establishing a test audio library according to the target reverberation signal by the testing module, wherein the test audio library includes test audio corresponding to the target reverberation signal; determining a target test audio from the test audio library by the testing module; and playing the target test audio by the testing module to test a target device.

[0007] An embodiment of the present application also provides a reverberation test system, including: a reverberation module, used to obtain reverberation audio according to original sound audio simulation, and generate a reverberation signal and an original sound signal; a processing module, used to process the reverberation signal to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets a preset similarity; a testing module, used to establish a test audio library according to the target reverberation signal, and the test audio library includes test audio corresponding to the target reverberation signal; the testing module is also used to determine the target test audio from the test audio library; the testing module is also used to play the target test audio to test the target device.

[0008] An embodiment of the present application also provides a computer device, including a memory storing a plurality of instructions; the processor loads instructions from the memory to execute the steps in any one of the reverberation testing methods provided in the embodiments of the present application.

[0009] The embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for a processor to load to execute the steps in any one of the reverberation testing methods provided in the embodiment of the present application.

[0010] In the embodiment of the present application, the reverberation module can obtain reverberation audio according to the original sound audio simulation, and generate a reverberation signal and an original sound signal; the processing module processes the reverberation signal to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets the preset similarity; the test module establishes a test audio library according to the target reverberation signal, and the test audio library includes test audio corresponding to the target reverberation signal; the test module determines the target test audio from the test audio library; the test module plays the target test audio to test the target device. In the present application, the reverberation audio can be obtained by simulating the original sound audio, and the target test audio for testing can be obtained by processing the reverberation audio, so that test audio applied to different reverberation scenarios can be generated, which is widely used in reverberation test scenarios and in the testing of various devices, thereby improving the efficiency of reverberation testing. At the same time, since the similarity between the obtained reverberation signal and the original sound signal meets the preset value, it is avoided that the distortion of the processed reverberation signal is too high, resulting in inaccurate reverberation test results, so the present application can also improve the accuracy of the reverberation test results. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1a is a schematic diagram of a scenario of a reverberation testing method provided in an embodiment of the present application;

[0013] Figure 1b is a flow chart of a reverberation testing method provided in an embodiment of the present application;

[0014] Figure 2 is a structural schematic diagram of a reverberation test system provided in an embodiment of the present application;

[0015] Figure 3 It is another flowchart of the reverberation testing method provided in the embodiment of the present application;

[0016] Figure 4 It is a structural schematic diagram of a silent shielding processing unit provided in an embodiment of the present application;

[0017] Figure 5 is a schematic diagram of the structure of a reverberation recording unit provided in an embodiment of the present application;

[0018] Figure 6 It is a structural schematic diagram of a surround playback unit provided in an embodiment of the application;

[0019] Figure 7 It is another structural schematic diagram of a surround playback unit provided in an embodiment of the application;

[0020] Figure 8 is another structural schematic diagram of the reverberation test system provided in an embodiment of the present application;

[0021] Fig. 9 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0023] Embodiments of the present application provide a reverberation testing method, system, computer device, and storage medium.

[0024] The reverberation test sample system can be integrated into an electronic device, which can be a terminal, a server, or other devices. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, or a personal computer (PC), etc. The server can be a single server or a server cluster composed of multiple servers.

[0025] In some embodiments, the reverberation test system may also be integrated into multiple electronic devices. For example, the reverberation test sample system may be integrated into multiple servers, and the reverberation test sample method of the present application may be implemented by multiple servers.

[0026] In some embodiments, the server may also be implemented in the form of a terminal.

[0027] For example, refer to Figure 1a In some embodiments, a scenario diagram of a reverberation test system is provided, and the performance test system may include a server 1000 and a network 2000. A reverberation test device is integrated in the server 1000, and the server 1000 is used to perform a performance test on a target device 3000. The network 2000 is used for data transmission between the server 1000 and the target device 3000, and the network 2000 may be a wireless network or a wired network, such as a wireless network such as a wireless local area network (WLAN), a local area network (LAN), a cellular network, a 2G network, a 3G network, a 4G network, a 5G network, etc. The target device 3000 may be a terminal, a server, or other equipment, and the server may be a single server or a server cluster consisting of multiple servers.

[0028] Among them, the server obtains reverberation audio according to the original sound audio simulation through the reverberation module, and generates a reverberation signal and an original sound signal; through the processing module, the reverberation signal is processed to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets the preset similarity; through the testing module, a test audio library is established according to the target reverberation signal, and the test audio library includes test audio corresponding to the target reverberation signal; through the testing module, the target test audio is determined from the test audio library; through the testing module, the target test audio is played to test the target device.

[0029] It should be noted that the serial numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0030] In this embodiment, a reverberation test method is provided. The reverberation test method is applied to a reverberation test system. The reverberation test system includes a reverberation module, a processing module, and a test module. Figure 1b As shown, the specific process of the reverberation test method can be as follows:

[0031] 110. Through the reverberation module, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated.

[0032] Among them, the reverberation module refers to a module that simulates the original sound to obtain the reverberation audio, which can be a reverberation effector, or a sound playing device set in a reverberation room, etc. The original sound audio refers to the audio used to obtain the reverberation audio, and the reverberation audio refers to the audio obtained by reverberating the original sound audio. The reverberation processing may include adjusting the decay time of the original sound audio, the delay time of the front and rear reflections, or the high and low frequency cutoff parameters through the reverberation effector, or the reverberation audio may be obtained by playing the original sound audio in the reverberation room. The reverberation signal and the original sound signal refer to the audio signals obtained by processing the reverberation audio and the original sound audio, respectively. The audio signal is a signal representing a mechanical wave, and is an information carrier of the wavelength and intensity changes of the mechanical wave. It should be noted that in the embodiment of the present application, when the intelligent voice recognition product is subjected to a reverberation test, the original sound audio is speech.

[0033] The reverberation module is used to perform reverberation processing on the original sound audio to obtain the reverberation audio. When the reverberation effector is used for reverberation processing, the original sound audio is input into the reverberation effector to obtain the original sound signal. After the reverberation processing, the reverberation signal is obtained, and the corresponding reverberation audio is obtained. When the sound playback device set in the reverberation room is used for reverberation processing, the original sound audio is played in the reverberation room to generate the reverberation audio, and the reverberation signal is obtained after recording. The original sound signal can be generated by collecting the original sound audio in the reverberation room, or the original sound signal can be obtained from the played original sound audio.

[0034] In some embodiments, by playing the original audio in the reverberation room, and collecting the reverberation audio and the original audio, a real reverberation scene is better simulated to improve the simulation authenticity of the reverberation voice scene. The reverberation module may include an original audio playback unit and a reverberation recording unit. The reverberation module is arranged in the reverberation room. The steps of obtaining the reverberation audio according to the original audio simulation by the reverberation module, and generating the reverberation signal and the original audio signal may include:

[0035] The original sound audio is played in the reverberation room through the original sound playback unit to simulate the reverberation audio;

[0036] The reverberation audio and the target original sound audio are obtained through the reverberation recording unit, and a reverberation signal and an original sound signal are generated.

[0037] Among them, the original sound playback device refers to a unit used to play the original sound audio, and may include at least one device with a playback function, such as a speaker, a television, a mobile phone, and the like. A reverberation chamber refers to a laboratory that can fully reflect sound energy on all boundaries and fully diffuse it therein, so as to form a diffusion field with uniform energy density everywhere and random distribution in all propagation directions. The original sound played in the reverberation chamber can produce a reverberation effect. The reverberation recording unit refers to a unit used to record the sound in the reverberation chamber, and may include at least one device with a sound recording function, such as a microphone, a recorder, and the like. The target original sound audio is the audio obtained based on the original sound audio. The original sound audio can be directly used as the target original sound audio, or it can be recorded in the reverberation chamber, and the like. The reverberation signal is the audio signal of the reverberation audio, and the original sound signal is the audio signal of the target original sound audio.

[0038] In some embodiments, the reverberation recording unit includes a plurality of reverberation receiving subunits, the reverberation audio is recorded in a reverberation room by the plurality of reverberation receiving subunits, and the target original sound audio is the original sound audio.

[0039] In some embodiments, when a speech with a reverberation effect is recognized during the use of an intelligent speech recognition product, the processed original sound signal is obtained by processing the reverberation signal, that is, it has a reverberation scene feature. Therefore, in order to better simulate a real reverberation scene, the reverberation audio and the original sound audio are obtained by the reverberation receiving subunit and the original sound receiving subunit, respectively. Specifically, the reverberation recording unit includes a plurality of reverberation receiving subunits and a plurality of original sound receiving subunits. The step of obtaining the reverberation audio and the original sound audio through the reverberation recording unit, and generating the reverberation signal and the original sound signal may include:

[0040] Acquire reverberation audio through a plurality of reverberation receiving sub-units, and generate a reverberation signal according to the reverberation audio;

[0041] The original sound audio is obtained through a plurality of original sound receiving sub-units, and an original sound signal is generated according to the original sound audio.

[0042] The reverberation receiving subunit refers to a unit for collecting reverberation audio, and may include at least one device with a sound collection function, such as a microphone, a recorder, etc. The original sound receiving subunit refers to a unit for collecting original sound audio, and may include at least one device with a function of obtaining original sound audio, and may be a device for collecting sound based on sound waves, such as a sound wave collector, etc. The process of the original sound forming reverberation after reflection in the reverberation chamber is a mixture of direct sound waves and reflected sound waves. The frequency and amplitude of the original sound will not change after reflection. Therefore, by collecting sound waves through a sound wave collector, the original sound audio can be obtained based on the collected sound waves.

[0043] 120. Process the reverberation signal through the processing module to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets a preset similarity.

[0044] Among them, the processing module refers to a module that processes the reverberation signal to obtain the target reverberation signal, and the processing process may include but is not limited to filtering, amplitude modulation, frequency modulation, etc. of the reverberation signal. Similarity refers to the degree of similarity of the audio, which can be determined by but not limited to at least one of the waveform, human voice feature value, environmental feature value, etc. The preset similarity refers to the similarity between the preset target reverberation signal and the original sound signal. Since the similarity between the obtained reverberation signal and the original sound signal meets the preset value, it is avoided that the distortion of the processed reverberation signal is too high, resulting in inaccurate reverberation test results, so the accuracy of the reverberation test results can be improved.

[0045] In some implementations, the processing unit is connected to the reverberation recording unit, and the reverberation recording unit sends the generated reverberation signal and the original sound signal to the processing unit.

[0046] In some implementations, in order to obtain target reverberation information of different intensities corresponding to different reverberation scenes to improve the efficiency of reverberation testing, the step of processing the reverberation signal through a processing module to obtain the target reverberation signal may include:

[0047] Through the processing module, the recursive threshold and convergence condition of the offline solution in the MCLP model are determined according to the preset reverberation scene to obtain the target MCLP model;

[0048] The target reverberation signal is obtained by processing the reverberation signal through the target MCLP model and the original acoustic signal.

[0049] Among them, the MCLP (Multichannel Linear Prediction) model refers to a model that performs dereverberation processing on the target reverberation signal based on the MCLP algorithm. The MCLP algorithm takes the input as a plurality of microphone reception signals and outputs a speech signal after removing the reverberation. The MCLP algorithm model in the embodiment of the present application can be an algorithm model of the prior art. In the MCLP model, the recursive threshold of the offline solution can determine the similarity between the output reverberation signal and the original sound signal, and the convergence condition can determine the reverberation intensity of the output reverberation signal. By adjusting the convergence condition of the MCLP model, the target reverberation signal with different reverberation intensities can be obtained, and the similarity between the reverberation signal and the original sound signal can be changed by adjusting the recursive threshold of the offline solution. Among them, the preset reverberation scene refers to a pre-set reverberation scene, which can include conference rooms, shopping malls, etc., and different reverberation scenes have different reverberation intensity requirements. Therefore, the reverberation intensity requirement of the preset reverberation scene is used to determine the recursive threshold and convergence condition of the offline solution in the MCLP model, so as to obtain a reverberation signal that matches the reverberation intensity requirement of the preset reverberation scene, and the similarity between the reverberation signal and the original sound signal satisfies the preset threshold.

[0050] The convergence condition refers to the condition for determining whether the MCLP model converges, and may be the number of convergences determined according to a preset test scenario. Due to different convergence degrees, the output reverberation signal waveform is different.

[0051] In the process of reverberation audio processing, different reverberation signals can be obtained by adjusting the convergence conditions, so that the same reverberation audio can be applied to multiple scenarios after processing. By adjusting the recursive threshold of the offline solution, the obtained target reverberation signal can be more accurate.

[0052] In some embodiments, the processing module includes an MCLP model. Before the step of determining the recursive threshold and convergence condition of the offline solution in the MCLP model according to a preset reverberation scene through the processing module to obtain the target MCLP model, it can also include: obtaining a preset reverberation scene through the processing module.

[0053] In some implementations, when comparing the obtained reverberation signal with the original sound signal for similarity, the reverberation signal and the original sound signal are adjusted to meet a preset similarity by adjusting the parameters of the reverberation signal processing process to improve the accuracy of the reverberation test result. The step of processing the reverberation signal through the target MCLP model and the original sound signal to obtain the target reverberation signal may include:

[0054] Process the reverberation signal through the target MCLP model to obtain a processed reverberation signal;

[0055] When the similarity between the processed reverberation signal and the original sound signal meets a preset similarity, the processed reverberation information is determined as a target reverberation signal;

[0056] When the similarity between the processed reverberation signal and the original sound signal does not meet the preset similarity, the recursive threshold of the offline solution in the target MCLP model is adjusted to obtain the adjusted target MCLP model, and the reverberation signal is processed based on the adjusted target MCLP model to obtain the processed reverberation signal.

[0057] The processed reverberation signal is a signal obtained by dereverberation processing by the target MCLP model.

[0058] In some embodiments, the MCLP model includes an MCLP algorithm and an EM algorithm, and the MCLP model solves an offline solution through the EM algorithm. The step of determining the recursive threshold and convergence condition of the offline solution in the MCLP model according to a preset reverberation scene through a processing module to obtain a target MCLP model may include: determining the recursive threshold and convergence condition of the offline solution in the MCLP model according to a preset reverberation scene to obtain a target EM algorithm. Processing the reverberation signal through the target MCLP model to obtain a processed reverberation signal may include: calculating a column vector through the MCLP algorithm and the reverberation signal; performing an offline solution on the column vector through the target EM algorithm until the convergence condition is met; and obtaining the processed reverberation signal.

[0059] The EM algorithm (Expectation-maximization algorithm) is an algorithm for finding the maximum likelihood estimate or maximum a posteriori estimate of parameters in a probability model, where the probability model depends on unobservable hidden variables.

[0060] In some embodiments, in order to increase the accuracy of the similarity comparison process, the original acoustic signal when performing the similarity comparison can be the original acoustic signal processed by the target MCLP model, or the original acoustic signal can be processed by the target MCLP model before the similarity comparison to obtain the processed original acoustic signal. When performing the similarity comparison, it is determined whether the similarity between the processed reverberation signal and the processed original acoustic signal meets the preset similarity.

[0061] In some implementations, the processed reverberation signal is sequentially matched with the original sound signal to calculate the similarity, and the preset similarity may be greater than 90%.

[0062] 130. Establish a test audio library according to the target reverberation signal through the test module, where the test audio library includes test audio corresponding to the target reverberation signal.

[0063] The test module refers to a module that uses a target reverberation signal for testing. After obtaining the target reverberation signal, a corresponding test audio for playing is obtained from the target reverberation signal, and the test audio is added to the test audio library. After repeatedly executing steps 110 to 130, multiple target reverberation signals can be obtained, and the test audio library can include multiple corresponding test audios.

[0064] In some embodiments, step 110 may include: obtaining a preset test scene; simulating reverberation audio according to the original sound audio and the preset test scene through the reverberation module, and generating a reverberation signal and an original sound signal. In this way, the parameters of the reverberation module can be adjusted according to different preset test scenes to simulate different reverberation audios, so in step 130, a test audio library including test audios corresponding to different preset test scenes can be obtained.

[0065] In some implementations, in step 120, different convergence conditions may be set according to different preset test scenarios to obtain a plurality of target reverberation signals corresponding to different preset test scenarios. Thus, in step 130, a test audio library including test audio corresponding to different preset test scenarios may be obtained.

[0066] 140. Determine target test audio from a test audio library through a test module.

[0067] The target test audio refers to the test audio used to test the target device determined from the test audio in the test audio library. The target test audio can be randomly selected from the test audio library, or can be determined according to the test requirements, such as the reverberation scene to be tested, etc.

[0068] In some implementations, in order to obtain test audio that matches the reverberation scene to be tested and increase the pertinence of the test process, the step of determining the target test audio from the test audio library through the test module may include:

[0069] Obtain the reverberation scene to be tested through the test module;

[0070] The test audio corresponding to the reverberation scene to be tested in the test audio library is determined as the target test audio.

[0071] Among them, the reverberation scene to be tested refers to the scene in which the reverberation test is to be performed on the target device, and the reverberation scene to be tested may include test parameters such as reverberation scene scale, scene type, sound source surround square, volume, tone, and number of plays. The reverberation scene to be tested may be determined by input from the tester, or may be selected from the reverberation scenes to be tested used in historical tests. In the process of testing, the reverberation scene to be tested may be represented in the form of a test case. The reverberation scene to be tested may be matched and calculated with the audio in the test audio library, and the one with the highest degree of match may be determined as the target test audio. In some embodiments, the reverberation scene to be tested may be matched and calculated with the preset test scene corresponding to the test audio, and the test audio with the highest degree of match may be determined as the target test audio.

[0072] 150. Play target test audio through the test module to test the target device.

[0073] The test unit may include a playback unit for playing the target test audio. The playback mode may be surround playback or non-surround playback, etc. The target device refers to the device to be tested for the reverberation scene, for example, the device to be tested for its speech recognition capability in the reverberation scene, such as a smart speaker, a smart TV, a smart phone, etc.

[0074] By playing the target test audio, a reverberation scene is constructed for the target device. The target device receives the played target test audio through the constructed reverberation scene and recognizes the target test audio. For example, when the target test audio is audio containing speech, the target device performs speech recognition after receiving the target test audio, and the speech recognition capability of the target device is judged according to the speech recognition result.

[0075] In some embodiments, by setting multiple sound-emitting units and multiple sound distortion units, the target test audio is played in a surround manner, which can better simulate the reverberation scene. In order to simulate more reverberation scenes, by selecting different types of sound distortion units, a variety of playback scene options for the same audio are provided. The test module may include multiple sound-emitting units for playing the target test audio and multiple sound distortion units for interfering with the target test audio. The multiple sound distortion units include at least one of a sound absorbing unit or a reflective unit. In some embodiments, the multiple sound-emitting units and the multiple sound distortion units are arranged at intervals along the circumference of the target device. Or, the multiple sound-emitting units and the multiple sound distortion units are evenly distributed around the target device.

[0076] Among them, the sound-emitting unit can be a device that can play audio, such as a speaker, a smart TV, a mobile phone, etc. The sound-absorbing unit refers to a unit used to absorb sound, which can be a sound-absorbing structure such as a sound-absorbing panel or a sound-absorbing soft bag. The reflecting unit refers to a unit used to reflect sound, which can be a reflecting structure such as a glass plate or a metal plate. Surround playback is achieved through the surround sound-emitting unit, and a sound distortion unit is added to absorb or reflect sound to simulate different scenes, such as reverberation scenes such as shopping malls or conference rooms. It should be noted that in order to simulate reverberation scenes, multiple sound-emitting units and multiple sound distortion units can be set indoors.

[0077] The reverberation test scheme provided by the embodiment of the present application can be applied in various reverberation test scenarios. For example, taking the reverberation test of intelligent speech recognition products as an example, it is applied to the reverberation test system, and the reverberation test system includes a reverberation module, a processing module and a test module, and the method includes: through the reverberation module, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated; through the processing module, the reverberation signal is processed to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets the preset similarity; through the test module, a test audio library is established according to the target reverberation signal, and the test audio library includes a test audio corresponding to the target reverberation signal; through the test module, the target test audio is determined from the test audio library; through the test module, the target test audio is played to test the target device. The scheme provided by the embodiment of the present application is adopted to obtain reverberation audio through original sound audio simulation, and the target test audio for testing is obtained by processing the reverberation audio, so that test audio applied to different reverberation scenarios can be generated, which is widely used in reverberation test scenarios and in the testing of various devices, thereby improving the efficiency of reverberation testing. At the same time, since the similarity between the obtained reverberation signal and the original sound signal meets the preset value, the distortion of the processed reverberation signal is avoided to be too high, resulting in inaccurate reverberation test results. Therefore, the present application can also improve the accuracy of the reverberation test results.

[0078] At the same time, the embodiment of the present application processes the target test audio based on the reverberation audio collected from the real reverberation scene, thereby improving the simulation authenticity of the reverberation speech scene. In the process of processing the reverberation audio, by adjusting the convergence conditions, different reverberation signals can be obtained, increasing the application of the same reverberation audio in diverse scenarios, and by adjusting the recursive threshold of the offline solution, the obtained target reverberation signal can be made more accurate.

[0079] The method described in the above embodiment will be further described in detail below.

[0080] In this embodiment, the method of the embodiment of the present application will be described in detail by taking the reverberation test of an intelligent speech recognition product as an example.

[0081] A reverberation test method is applied to a reverberation test system, such as Figure 2 As shown, the reverberation test system may include a reverberation module, a processing module, a test module, and a master control and cloud host load cluster module, wherein the reverberation module may include a hemispherical reverberation scene recording unit, a recording control unit, and a mute shielding processing unit, and the test module may include a service input unit, a surround playback unit, a playback control unit, and a result output unit.

[0082] Among them, the hemispherical reverberation scene recording unit is responsible for recording the reverberation scene of the real world, and provides the function of all-round recording of sound radiating to the three-dimensional space based on the plane; the recording control unit is responsible for the switching and selection of recording tools (microphone, sound wave collector); the mute shielding processing unit is responsible for adjustable sound insulation; the processing module is used to perform WPE algorithm processing on the recorded audio; the business input unit is used to obtain the original audio, as well as the setting of the working status of the input hemispherical reverberation scene recording unit and the mute status setting of the reverberation room, and is also used to write test cases, Set the test time, test plan, select the audio data samples required for the use case, execute the test, and display the test information returned by the result output module unit; the surround playback unit, composed of multiple stereo speakers, is used to achieve surround playback; the playback control unit is used to control the start and stop, volume, tone, and balance of the surround playback unit speakers; the result output unit is used to integrate the test results and transmit them to the result dashboard display stand of the business input unit; the main control and cloud host load cluster module is responsible for loading the test plan content, scheduling database information, deploying processing modules, and performing load balancing scheduling of processing module operations.

[0083] For incomplete descriptions of the steps in this embodiment, please refer to the previous method embodiments and will not be repeated here.

[0084] like Figure 3 As shown, the specific process of the reverberation test method is as follows:

[0085] 210. Determine the working status of the original audio, the hemispherical reverberation scene recording unit, and the mute status of the reverberation room through the business input unit.

[0086] The service input unit can be used to obtain the original audio, as well as to obtain the working state setting of the input hemispherical reverberation scene recording unit and the mute state setting of the reverberation room.

[0087] 220. Adjust the working state of the hemispherical reverberation scene recording unit through the recording control unit.

[0088] The recording control unit is mainly responsible for switching and selecting recording tools (sound wave collector, microphone). The recording control unit adjusts the working state of the hemispherical reverberation scene recording unit to the working state determined by the business input unit.

[0089] Among them, the switching function can be used to enable the recording control unit to switch different modes through the script configuration file, and the recording control unit is preset with a sound wave collector identifier and a microphone identifier. For example, to switch the silent mode: call the control script, connect to the embedded switch device, and operate the sound gate of the silent shielding processing unit by powering off and on, so as to realize the switching between the electrically controlled silent environment and the non-silent environment. To switch the start and stop of the sound wave collector: call the control script, and according to the preset sound wave collector identifier, remotely switch the recording function of the sound card inserted in the device where the sound wave collector is located. To switch the start and stop of the microphone: call the control script, and according to the preset microphone identifier, remotely switch the recording function of the sound card inserted in the device where the microphone is located.

[0090] Among them, the selection function can be used to select and adjust the parameters of the device. According to the preset requirements given by the business input unit, the parameters of the scene type (concert hall / conference room / shopping mall, etc.), sound source surround direction (front / rear / left / right / top / various combinations of [front / rear / left / right / top / ], manual input is supported), volume, tone size (0~854 / Mel), and balance value (BAL left-right balance / FAD front-to-back balance) are transmitted. The parameters transmitted by the script and the underlying control logic call the console interface of the sound card for implementation.

[0091] 230. Adjust the mute state of the reverberation room through the mute shielding processing unit.

[0092] The recording control unit adjusts the working state of the mute shielding processing unit to the working state determined by the business input unit. Figure 4 As shown, the silent shielding processing unit can be composed of a general reverberation acoustic solution, the silent shielding processing unit can include a sound gate, the reverberation chamber includes an outer room and an inner room, and the silent shielding processing unit can be used for adjustable sound insulation between the outer room and the inner room. For example, the recording control unit adjusts the silent state of the reverberation chamber by controlling the sound gate: the recording control unit calls the control script, connects to the embedded switch device, and operates the sound gate of the silent shielding processing unit in a power-off and power-on manner to achieve the switching between the electrically controlled silent environment and the non-silent environment. The material selection of the reverberation chamber meets the high-efficiency sound absorption structure. Above the cutoff frequency, the sound absorption coefficient is greater than 0.99. The enclosure structure of the reverberation chamber should be thick, the inner surface should be hard and smooth, and the interface sound absorption coefficient should be very small, so that most of the indoor space meets the conditions of the diffuse sound field. The body shape is approximately rectangular. In order to increase the diffusion degree of the indoor sound field, the inner surface is made into a cylindrical or other convex body. Materials with very small sound absorption coefficients such as tiles, marble, and terrazzo are used, and the average sound absorption coefficient does not exceed 0.06.

[0093] 240. Through the hemispherical reverberation scene recording unit, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated.

[0094] The hemispherical reverberation scene recording unit provides a function of recording the sound radiated to the three-dimensional space in all directions based on a plane. The hemispherical reverberation scene recording unit may include an original sound playback unit, a reverberation room, and a reverberation recording unit.

[0095] The reverberation recording unit includes a plurality of reverberation receiving subunits and a plurality of original sound receiving subunits. For example, the reverberation recording unit may be composed of 5 microphones (i.e., reverberation receiving subunits) and 4 sound wave collectors (i.e., original sound receiving subunits). Figure 5 As shown, Figure 5 For a bird's-eye view, five microphones are deployed in front, behind, left, right, and above the object being recorded (i.e., the original sound playback unit). The audio cables of the five microphones are connected to five hosts with sound cards through professional XLR (large three-core, the purpose is to eliminate interference) interfaces. The five sound cards capture the scene sound, and the captured audio information is not filtered. Therefore, the recorded sound is equivalent to the original sound of the scene of the recording object.

[0096] The four sound wave collectors are placed on the four horizontal lines of the recorded object, and the directions are respectively diagonal lines of the xyz axis (positive diagonal X direction) for sound wave recording. The sound broadcast recording is connected to the host sound card audio inlet in two directions through the TRRS connector (small four-core) and the audio 1-input 2-output expansion port. A simple single-sided sound barrier is set up about 1 to 5 meters in front of the sound wave collector. The sound barrier can filter the high-frequency sound emitted by the recorded scene object. The purpose of setting it up is to enable the sound wave collector to accurately collect the original sound of the recorded object. The functions of the four sound wave collectors are: to filter out the volume exceeding 85 decibels (harmful to the human ear); and to reduce unnecessary background noise. There is a noise reduction knob on the side of the sound wave collector, which can be adjusted to a suitable noise reduction frequency to make the listening sound clearer for subsequent processing modules to process.

[0097] 250. Process the reverberation signal through the processing module to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets a preset similarity.

[0098] The processing module uses the MCLP algorithm to process the audio recorded by the reverberation recording unit. The MCLP algorithm is one of the WPE (Weighted Prediction Error) algorithms. The voice signals received by 5 professional microphones and 4 sound wave collectors are input into the processing module, and the voice signals after reverberation (voice tailing) are removed after processing. The voice signal itself is one-dimensional, which is the characteristic of the sound itself. For scientific analysis, the voice signal needs to be short-time Fourier transformed in symbolic representation, and finally presented as a waveform diagram. Before the voice signal is processed, the incoming voice is equivalent to the original voice of the object being recorded directly received by the microphone / sound wave collector. The data is played back through the sound card on the computer as waveform data (i.e., simulated waveform), so there will be parameters such as frame, frequency, time, and signal size. The meaning of the WPE algorithm is weighted prediction error, which is a broad audio processing algorithm, and the MCLP algorithm is one of them, which means multi-channel linear prediction, that is, an optimization processing algorithm that supports multiple microphones / sound wave collectors to remove reverberation noise under common input.

[0099] The MCLP model can represent the current received signal as a linear combination of the current pure signal and the received signals of several frames in the past. Ignoring the noise, it can be expressed as follows:

[0100] Where n is the frame number, l is the frequency, M is the number of microphones, and y is 1，n，l represents the signal of the first microphone at the lth frequency point in the nth frame. Therefore, with the number of microphones as the vector dimension, the received signals of multiple microphones are written in vector form as y n，l =(y 1，n，l ,y 2，n，l , ..., y M，n，l ) T . The distance between frame numbers is expressed as the vector dimension: g l is a column vector of length MK, called the linear coefficient. The purpose of the dereverberation algorithm is to estimate g l .s n，l Represents the pure signal, which is obtained by the original signal y 1，n，l Subtract reverberation noise Got it.

[0101] Where solving for g l Use the following method:

[0102] (1) Objective function

[0103]

[0104] Solving for g lThe idea is to maximize the likelihood function. Assuming that the received signals of n=1~N frames are known, the optimization problem can be expressed as:

[0105]

[0106] In the formula, And assume that s n，l It obeys the time-varying complex Gaussian distribution, that is, s n，l ~N C (0, λ n，l ), substituting the probability expression of Gaussian distribution into the above formula, it can be seen that the above optimization problem is equivalent to .

[0107] (2) Offline solution

[0108] The above objective function contains two sets of parameters g l and λ n，l . The EM algorithm can be used to solve it. First, fix g l Invariant solution λ n，l ,have Fix λ again n，l Unchanged solution g l ,Right now The above objective function is g l The quadratic function of g l Taking the derivative and setting it to zero gives the analytical solution:

[0109] Repeat the above formula to calculate g l and λ n，l Until convergence. Then g l The analytical solution of is often called an offline solution.

[0110] In the processing module, the recursive threshold of the offline solution can be set according to the scene requirements (such as high reverberation intensity, medium reverberation intensity, and low reverberation intensity) (i.e., repeatedly calculating g and λ in the above formula until convergence), and the number of convergence calculations can be given according to the test scene requirements, so that the processed audio can present outputs with different reverberation effects. Due to different degrees of convergence, the output signal waveform results are different. When the number of microphones M is changed (the default value of M in this embodiment is 5), the processing module can also dynamically adjust M according to the start and stop state result parameters returned by the recording control unit.

[0111] After obtaining the adjusted audio through the above process, the adjusted audio and the audio collected by the sound wave collector can be matched sequentially. If the similarity is greater than 90%, it is marked as a successful optimization. Otherwise, the background record prompts the audio optimization result: failed, and adjusts the recursive threshold for correction.

[0112] The master control and cloud host load cluster module is responsible for loading the test plan content, scheduling database information, deploying processing modules, and performing load balancing scheduling of processing module operations. Among them, loading the test plan content can obtain the suspended test plan of the business input unit for the main control code, and perform configuration file installation and parameterization through the test case set associated with the test plan, and trigger the function of calling the database access. Scheduling database information can be used for loading the case, and the acquisition of test resources will be executed due to the triggering of the function of calling the database access, including but not limited to: test files, audio materials, etc.; the WPE operation results of the processing module are cached through the master control and cloud host load cluster module, and the database is called for insertion to store audio data samples for testing. Deploy the processing module and perform load balancing scheduling of the processing module operation: distribute and replicate the processing module operation scripts on multiple cloud hosts, and the load balancing logic adopts the K8S balancing strategy.

[0113] 260. Obtain the test case through the business input unit, determine the working status of the target test audio and the surround playback unit according to the test case, and load the test case and the target test audio through the main control and cloud host load cluster unit.

[0114] The business input unit is responsible for supporting test engineers in writing and storing test cases, setting test time, formulating and issuing test plans, selecting audio data samples required for the test cases, triggering test execution, and displaying the test information returned by the result output unit.

[0115] Among them, writing and storing test cases. In the early stage of intervening in the reverberation scene voice test, it is necessary to write complete test cases, which need to include: reverberation scene scale (large / medium / small), scene type (concert hall / conference room / shopping mall, etc.), whether it is silent mode (yes / no), sound source surround direction (various combinations of front / back / left / right / up, support manual input), volume, tone size (0~854 / Mel), balance value (BAL left-right balance / FAD front-back balance), number of plays (1~10000), play interval (1~300s) and other scene test points. Test engineers can write and debug test cases on the test server and store the finalized test cases. Set the test time, which can be triggered according to the date (by calling the Jenkins integration framework to trigger the timer), that is: year-month-day-hour-minute-second, and can be ended according to the date. Even if the test is not completed, it will be triggered to end. Formulate and issue test plans. The test plans that can be formulated are composed of test cases, test time, and executors. Before issuing, a data integrity (filling integrity) check will be performed. The check is triggered by the test engineer when clicking Save; when clicking Run, the elements of the test plan will be transmitted to the master control and cloud host load cluster module for loading. Select the audio data samples required for the use case: through the matching algorithm, automatically select from the audio sample database according to the parameters of the test case. The result of the selection is that when the test plan is triggered to execute, it is loaded according to the associated test case. Trigger execution test: that is, the function of starting the test execution can be triggered by clicking the Execute Now button on the test plan page, or it can be automatically triggered without manual clicks in the above-mentioned set test time state, or under the dashboard where the test time state has been set and the service is suspended, click Execute Now to trigger the test execution. Display the test information returned by the result output unit: organize the test report in a predetermined format, store the report, and send the test report to the relevant objects.

[0116] 270. Adjust the working state of the surround playback unit through the playback control unit.

[0117] The playback control unit adjusts the working state of the surround playback unit to the working state determined by the business input unit. The playback control unit is used to control the start and stop, volume, tone, and balance of the surround playback unit. The control principle is consistent with that of the recording control unit, that is, preset the audio identifier (such as audio 1, audio 2) through the script configuration file. Call the control script and pass the parameters of the surround direction of the sound source, volume, tone (0~854 / Mel), and balance value (BAL left-right balance / FAD front-to-back balance) according to the preset requirements of the business input unit. The parameters passed by the script and the underlying control logic call the console interface of the sound card for implementation.

[0118] 280. Play the target test audio through the surround playback unit to test the intelligent voice recognition product.

[0119] The surround playback unit may include multiple sound generating units and multiple sound distortion units. Figures 6-7 As shown, the surround playback unit is composed of multiple stereo speakers, which can realize surround playback. The intelligent voice recognition product is a smart TV.

[0120] like Figure 6 As shown in the figure, the surround playback unit consists of 4 stereo speakers and four sound-absorbing structures located at the 4 corners of the wall. The speakers are hung in the middle of the wall instead of on the floor. The characteristics of this solution: filtering the reflection from the corners to cause partial distortion of the original sound and then transmitting it to the object under test, simulating a highly original sound scene.

[0121] like Figure 7 As shown, the surround playback unit consists of 4 independent stereo speakers. The speakers are hung in the middle of the corners instead of on the floor. The feature of this solution is that the original sound is emitted from the four corners and then reflected on the wall, simulating a high noise situation.

[0122] The above two options can be selected based on the reverberation scene scale (large / medium / small) and scene type (concert hall / conference room / shopping mall, etc.) selected by the test case: If the reverberation scale is large and the scene type is a shopping mall, select Figure 7 If it is a simulated conference room, select Figure 6 The selected control is selected by a script call in the playback control unit.

[0123] 290. Obtain and integrate test results through the result output unit.

[0124] The result output unit is responsible for data integration of the test results and transmits them to the result display stand of the business input unit. The test results are the results of testing the intelligent speech recognition product using the target test audio.

[0125] Among them, the transmission function can be that the original data of the test report of the business input unit comes from the result output unit, and the data is transmitted by network transmission, which specifically includes: data communication with the business input unit through network connection; data stream transmission, the result calculated by data integration is sent to the business input unit in the form of a file for its analysis, and the file identifier is accompanied by the test version, test time, and other basic test information. The data integration function can include statistics, calculation of test pass rate, use case coverage, problem summary, and automatic generation of DI value. DI value = number of fatal level problems * 10 + number of serious level problems * 3 + number of general level problems * 1 + number of prompt level problems * 0.1. These calculation results will be temporarily stored on the host, and at the same time, they will be triggered by the background service to perform real-time backup (establish database connection, upload) without shutting down the host to prevent power-off information loss, specifically including: the script obtains the status performance log of the device under test (target device) and the control log of the playback control unit; the script executes the calculation of the above test result indicators; temporarily stores the host locally for data transmission, and uploads the database in real time for storage.

[0126] As can be seen from the above, this embodiment obtains reverberation audio through original sound audio simulation, and obtains target test audio for testing through reverberation audio, which can be widely used in the testing of various devices to improve test efficiency. At the same time, the embodiment of the present application is based on the reverberation audio obtained by real reverberation scene collection, recording technology, post-recording processing technology, sound splitting and surround sound playback, and finally processing to obtain target test audio, which improves the simulation authenticity of reverberation voice scene. In the process of processing reverberation audio, by adjusting the convergence conditions, different reverberation signals can be obtained, and the diversity of scene applications for the same reverberation audio is increased. By adjusting the recursive threshold of the offline solution, the obtained target reverberation signal is more accurate. At the same time, during the test process, the target test audio is automatically selected according to the reverberation scene test case, the test is executed and the test report is generated, which improves the efficiency of reverberation testing.

[0127] In order to better implement the above method, the embodiment of the present application also provides a reverberation test system, and the reverberation test device can be integrated in an electronic device, and the electronic device can be a terminal, a server, etc. Among them, the terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop, a personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers.

[0128] For example, in this embodiment, the method of the embodiment of the present application will be described in detail by taking the specific integration of the reverberation test system in the server as an example.

[0129] For example, Figure 8As shown, the reverberation test system may include a reverberation module 310, a processing module 320 and a testing module 330, as follows:

[0130] (I) Reverberation module

[0131] It is used to obtain reverberation audio according to the original audio simulation, and generate reverberation signal and original audio signal.

[0132] In some embodiments, the reverberation module includes an original sound playback unit and a reverberation recording unit. The reverberation module is disposed in a reverberation room. The reverberation module can be specifically used for:

[0133] The original sound audio is played in the reverberation room through the original sound playback unit to simulate the reverberation audio;

[0134] The reverberation audio and the target original sound audio are obtained through the reverberation recording unit, and a reverberation signal and an original sound signal are generated.

[0135] In some implementations, the reverberation recording unit includes a plurality of reverberation receiving subunits and a plurality of original sound receiving subunits, and the reverberation recording unit may be specifically used for:

[0136] Acquire reverberation audio through a plurality of reverberation receiving subunits, and generate a reverberation signal according to the reverberation audio;

[0137] The original sound audio is obtained through a plurality of original sound receiving sub-units, and an original sound signal is generated according to the original sound audio.

[0138] (II) Processing Module

[0139] It is used to process the reverberation signal to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets a preset similarity.

[0140] In some embodiments, the processing module may be specifically used to:

[0141] Through the processing module, the recursive threshold and convergence condition of the offline solution in the MCLP model are determined according to the preset reverberation scene to obtain the target MCLP model;

[0142] The target reverberation signal is obtained by processing the reverberation signal through the target MCLP model and the original acoustic signal.

[0143] In some implementations, processing the reverberation signal through the target MCLP model and the original acoustic signal to obtain the target reverberation signal may include:

[0144] Process the reverberation signal through the target MCLP model to obtain a processed reverberation signal;

[0145] When the similarity between the processed reverberation signal and the original sound signal meets a preset similarity, the processed reverberation information is determined as a target reverberation signal;

[0146] When the similarity between the processed reverberation signal and the original sound signal does not meet the preset similarity, the recursive threshold of the offline solution in the target MCLP model is adjusted to obtain the adjusted target MCLP model, and the reverberation signal is processed based on the adjusted target MCLP model to obtain the processed reverberation signal.

[0147] (III) Test module

[0148] The test module is used to establish a test audio library according to the target reverberation signal, wherein the test audio library includes test audio corresponding to the target reverberation signal. The test module can also be used to: determine the target test audio from the test audio library. The test module can also be used to: play the target test audio to test the target device.

[0149] In some implementations, determining a target test audio from a test audio library may include:

[0150] Obtain the reverberation scene to be tested through the test module;

[0151] The test audio corresponding to the reverberation scene to be tested in the test audio library is determined as the target test audio.

[0152] In some embodiments, the test module includes multiple sound-emitting units for playing target test audio and multiple sound distortion units for interfering with the target test audio. The multiple sound-emitting units and the multiple sound distortion units are spaced around the target device, and the multiple sound distortion units include at least one of a sound absorbing unit or a reflecting unit.

[0153] In specific implementation, the above units can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units can refer to the previous method embodiments, which will not be repeated here.

[0154] Therefore, the embodiment of the present application can obtain reverberation audio through original sound audio simulation, and obtain target test audio for testing by processing the reverberation audio. It can generate test audio for different reverberation scenarios, and is widely used in reverberation test scenarios and in the testing of various devices, thereby improving the efficiency of reverberation testing.

[0155] Correspondingly, an embodiment of the present application also provides a computer device, which may be a terminal or a server, and the terminal may be a smart phone, a tablet computer, a laptop computer, a touch screen, a game console, a personal computer, a personal digital assistant (PDA), or other terminal devices.

[0156] like Fig. 9 As shown, Fig. 9 The schematic diagram of the structure of the computer device provided in the embodiment of the present application is that the computer device 400 includes a processor 410 having one or more processing cores, a memory 420 having one or more computer-readable storage media, and a computer program stored in the memory 420 and executable on the processor. The processor 410 is electrically connected to the memory 420. It can be understood by those skilled in the art that the computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0157] The processor 410 is the control center of the computer device 400. It uses various interfaces and lines to connect various parts of the entire computer device 400, executes various functions of the computer device 400 and processes data by running or loading software programs and / or modules stored in the memory 420, and calling data stored in the memory 420, thereby monitoring the computer device 400 as a whole.

[0158] In the embodiment of the present application, the processor 410 in the computer device 400 will load instructions corresponding to the processes of one or more application programs into the memory 420 according to the following steps, and the processor 410 will run the application programs stored in the memory 420 to implement various functions:

[0159] Through the reverberation module, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated; through the processing module, the reverberation signal is processed to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets the preset similarity; through the test module, a test audio library is established according to the target reverberation signal, and the test audio library includes test audio corresponding to the target reverberation signal; through the test module, the target test audio is determined from the test audio library; through the test module, the target test audio is played to test the target device.

[0160] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0161] Optional, such as Fig. 9 As shown, the computer device 400 further includes: a touch screen 430, a radio frequency circuit 440, an audio circuit 450, an input unit 460, and a power supply 470. The processor 410 is electrically connected to the touch screen 430, the radio frequency circuit 440, the audio circuit 450, the input unit 460, and the power supply 470, respectively. Those skilled in the art will appreciate that Fig. 9The computer device structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0162] The touch display screen 430 can be used to display a graphical user interface and receive operation instructions generated by the user acting on the graphical user interface. The touch display screen 430 may include a display panel and a touch panel. Among them, the display panel may be used to display information input by the user or information provided to the user and various graphical user interfaces of the computer device, which may be composed of graphics, text, icons, videos and any combination thereof. Optionally, the display panel may be configured in the form of a liquid crystal display (LCD, Liquid Crystal Display), an organic light-emitting diode (OLED, Organic Light-Emitting Diode) and the like. The touch panel may be used to collect the user's touch operation on or near it (such as the user using any suitable object or accessory such as a finger, stylus, etc. on the touch panel or near the touch panel), and generate corresponding operation instructions, and the operation instructions execute corresponding programs. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into the touch point coordinates, and then sends it to the processor 410, and can receive the command sent by the processor 410 and execute it. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it is transmitted to the processor 410 to determine the type of touch event, and then the processor 410 provides a corresponding visual output on the display panel according to the type of touch event. In an embodiment of the present application, the touch panel and the display panel can be integrated into the touch display screen 430 to realize the input and output functions. However, in some embodiments, the touch panel and the display panel can be used as two independent components to realize the input and output functions. That is, the touch display screen 430 can also be used as a part of the input unit 460 to realize the input function.

[0163] The radio frequency circuit 440 may be used to send and receive radio frequency signals, so as to establish wireless communication with a network device or other computer devices through wireless communication, and to send and receive signals between the network device or other computer devices.

[0164] The audio circuit 450 can be used to provide an audio interface between the user and the computer device through a speaker and a microphone. The audio circuit 450 can transmit the electrical signal converted from the received audio data to the speaker, which is converted into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 450 and converted into audio data, and then the audio data is output to the processor 410 for processing, and then sent to another computer device through the radio frequency circuit 440, or the audio data is output to the memory 420 for further processing. The audio circuit 450 may also include an earphone jack to provide communication between an external headset and the computer device.

[0165] The input unit 460 may be used to receive input numbers, character information or user feature information (such as fingerprint, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0166] The power supply 470 is used to supply power to various components of the computer device 400. Optionally, the power supply 470 can be logically connected to the processor 410 through a power management system, so that the power management system can manage charging, discharging, and power consumption. The power supply 470 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0167] although Fig. 9 Not shown, the computer device 400 may also include a camera, a sensor, a wireless fidelity module, a Bluetooth module, etc., which will not be described in detail here.

[0168] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0169] From the above, it can be seen that the computer device provided in this embodiment can obtain reverberation audio through original sound audio simulation, and obtain target test audio for testing by processing the reverberation audio. It can generate test audio for different reverberation scenarios, and is widely used in reverberation test scenarios and in the testing of various devices, thereby improving the efficiency of reverberation testing.

[0170] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0171] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a plurality of computer programs are stored, and the computer program can be loaded by a processor to execute the steps in any of the reverberation test methods provided in the embodiments of the present application. For example, the computer program can execute the following steps:

[0172] Through the reverberation module, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated; through the processing module, the reverberation signal is processed to obtain a target reverberation signal, and the similarity between the target reverberation signal and the original sound signal meets the preset similarity; through the test module, a test audio library is established according to the target reverberation signal, and the test audio library includes test audio corresponding to the target reverberation signal; through the test module, the target test audio is determined from the test audio library; through the test module, the target test audio is played to test the target device.

[0173] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0174] The storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0175] Since the computer program stored in the storage medium can execute the steps in any one of the reverberation testing methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the reverberation testing methods provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0176] The above is a detailed introduction to a reverberation test method, system, computer device and storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for technicians in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present application.

Claims

1. A reverberation testing method, characterized in that: Applied to a reverberation test system, the reverberation test system includes a reverberation module, a processing module and a test module, and the method includes: Through the reverberation module, reverberation audio is obtained according to the original sound audio simulation, and a reverberation signal and an original sound signal are generated; The reverberation signal is processed by the processing module to obtain a target reverberation signal, wherein the similarity between the target reverberation signal and the original sound signal satisfies a preset similarity; By means of the test module, a test audio library is established according to the target reverberation signal, wherein the test audio library includes test audio corresponding to the target reverberation signal; Determining target test audio from the test audio library through the test module; Play the target test audio through the test module to test the target device; The processing module processes the reverberation signal to obtain a target reverberation signal, including: By means of the processing module, a recursive threshold and a convergence condition of an offline solution in the MCLP model are determined according to a preset reverberation scene to obtain a target MCLP model; Processing the reverberation signal through a target MCLP model to obtain a processed reverberation signal; When the similarity between the processed reverberation signal and the original sound signal satisfies a preset similarity, determining the processed reverberation information as a target reverberation signal; When the similarity between the processed reverberation signal and the original sound signal does not meet the preset similarity, the recursive threshold of the offline solution in the target MCLP model is adjusted to obtain the adjusted target MCLP model, and the reverberation signal is processed based on the adjusted target MCLP model to obtain the processed reverberation signal.

2. The reverberation testing method according to claim 1, characterized in that: The reverberation module includes an original sound playback unit and a reverberation recording unit. The reverberation module is arranged in a reverberation room. The reverberation module obtains reverberation audio according to the original sound audio simulation, and generates a reverberation signal and an original sound signal, including: Playing the original sound audio in the reverberation chamber through the original sound playing unit to simulate the reverberation audio; The reverberation audio and the target original sound audio are obtained through the reverberation recording unit, and a reverberation signal and an original sound signal are generated.

3. The reverberation testing method according to claim 2, characterized in that: The reverberation recording unit includes a plurality of reverberation receiving subunits and a plurality of original sound receiving subunits. The reverberation recording unit is used to obtain the reverberation audio and the original sound audio, and to generate a reverberation signal and an original sound signal, including: Acquire the reverberation audio through the multiple reverberation receiving subunits, and generate a reverberation signal according to the reverberation audio; The original sound audio is acquired through the multiple original sound receiving sub-units, and an original sound signal is generated according to the original sound audio.

4. The reverberation testing method according to claim 1, characterized in that: Determining the target test audio from the test audio library by the test module includes: Obtaining the reverberation scene to be tested through the test module; The test audio corresponding to the reverberation scene to be tested in the test audio library is determined as the target test audio.

5. The reverberation testing method according to any one of claims 1 to 4, characterized in that: The test module includes multiple sound-emitting units for playing the target test audio and multiple sound distortion units for interfering with the target test audio. The multiple sound-emitting units and the multiple sound distortion units are arranged at intervals along the circumference of the target device, and the multiple sound distortion units include at least one of a sound absorbing unit or a reflecting unit.

6. A reverberation test system, characterized in that: include: A reverberation module, used to obtain reverberation audio according to the original audio simulation, and generate a reverberation signal and an original audio signal; A processing module, used for processing the reverberation signal to obtain a target reverberation signal, wherein the similarity between the target reverberation signal and the original sound signal satisfies a preset similarity; A testing module, used to establish a test audio library according to the target reverberation signal, wherein the test audio library includes test audio corresponding to the target reverberation signal; The test module is further used to determine the target test audio from the test audio library; The test module is further used to play the target test audio to test the target device; The processing module is further used to determine the recursive threshold and convergence condition of the offline solution in the MCLP model according to a preset reverberation scene to obtain a target MCLP model; process the reverberation signal through the target MCLP model to obtain a processed reverberation signal; when the similarity between the processed reverberation signal and the original sound signal meets the preset similarity, determine the processed reverberation information as the target reverberation signal; when the similarity between the processed reverberation signal and the original sound signal does not meet the preset similarity, adjust the recursive threshold of the offline solution in the target MCLP model to obtain an adjusted target MCLP model, and process the reverberation signal based on the adjusted target MCLP model to obtain a processed reverberation signal.

7. A computer device, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps in the reverberation testing method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the reverberation testing method according to any one of claims 1 to 5.

Images