Acoustic quantitative evaluation method and device for vibration performance of equipment motor

Abstract

The invention provides an acoustic quantitative evaluation method and device for the vibration performance of an equipment motor, and relates to the technical field of mobile equipment performance test and evaluation, and the method comprises the steps: starting a vibration motor of tested equipment, and obtaining an original audio signal generated when the vibration motor works; preprocessing the original audio signal to obtain a preprocessed acoustic signal; based on the preprocessed acoustic signals, acoustic characteristic parameters are extracted, and characteristic vectors are constructed based on the acoustic characteristic parameters; and inputting the feature vector into a pre-trained vibration performance evaluation model, and outputting a quantitative evaluation result of the performance of the vibration motor by comparing the feature vector with preset standard feature data. Accurate mapping between vibration performance and objective data is established by collecting acoustic signals of motor vibration and extracting quantifiable characteristic parameters. A pre-trained machine learning model is used for automatic matching and grading, a consistent quantitative result is output, human differences are eliminated, and a reliable and uniform standard is provided for product quality control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mobile device performance testing and evaluation, and particularly relates to an acoustic quantitative evaluation method and device for motor vibration performance of a device. BACKGROUND

[0002] With the development of mobile communication technology, electronic devices such as smart phones and tablet computers have widely integrated vibration motors for providing haptic feedback. The performance parameters of motor vibration, such as intensity, frequency and response delay, directly determine the quality of the user's haptic experience. Therefore, it is crucial to accurately and objectively evaluate the motor vibration performance in the device research and development, production testing and quality control links.

[0003] At present, the evaluation of motor vibration performance in the industry mainly relies on the following technical solutions: subjective hand feeling evaluation method: this method relies on the test personnel to hold the device, and the vibration intensity after the motor starts is felt by the human body, and the level is determined according to experience or grading standard. This scheme is usually applied to the research and development debugging stage, but it has inherent defects such as low efficiency, strong subjectivity, different evaluation results for different people, poor consistency and comparability, and cannot meet the standardized and quantitative testing needs. Laser vibration measurement method: this scheme uses professional equipment such as laser Doppler vibration meter to non-contact scanning measurement of the micro-vibration of the motor or its vibration conduction components (such as the device middle frame), which can obtain high-precision vibration frequency spectrum and acceleration data. Although this method has high measurement accuracy, the required equipment is extremely expensive and the operation is complex, which cannot be deployed on the mass production line or consumer terminal device, and is only limited to laboratory environment. Acceleration sensor acquisition method: this scheme pastes a micro-electro-mechanical system acceleration sensor on the mainboard or the shell near the motor to collect the vibration acceleration waveform of the motor when it works, and then analyzes the frequency spectrum energy characteristics through fast Fourier transform algorithm. This method can provide objective data, but it cannot completely evaluate the overall performance of the vibration energy transmitted to the user's handheld part through the whole machine structure, which deviates from the actual user experience.

[0004] In summary, the existing motor vibration performance evaluation technology has obvious limitations, therefore, there is an urgent need for an evaluation method that can utilize the device's own hardware resources, realize low-cost, automation, quantification and indirectly reflect the whole machine vibration conduction effect, to improve the testing efficiency, consistency and practicability, and meet the needs of large-scale production and quality control. SUMMARY

[0005] The present application provides an acoustic quantitative evaluation method for motor vibration performance of a device. The main purpose is to solve the problems of unstable results, complex deployment, low testing efficiency and inability to effectively evaluate the whole machine vibration conduction performance of the existing motor vibration performance evaluation technology.

[0006] According to a first aspect of the present application, a method for acoustically quantitatively evaluating the vibration performance of a device motor is provided, comprising:

[0007] starting a vibration motor of a device under test, and obtaining an original audio signal generated by the vibration motor; preprocessing the original audio signal to obtain a preprocessed acoustic signal; extracting acoustic feature parameters based on the preprocessed acoustic signal, and constructing a feature vector based on the acoustic feature parameters; inputting the feature vector into a pre-trained vibration performance evaluation model, comparing the feature vector with preset standard feature data, and outputting a quantitative evaluation result of the vibration motor performance.

[0008] Preferably, the preprocessing of the original audio signal to obtain a preprocessed acoustic signal comprises time-frequency transformation and noise reduction filtering processing of the original audio signal to obtain a pure time-frequency domain acoustic signal.

[0009] Preferably, the extraction of acoustic feature parameters based on the preprocessed acoustic signal and the construction of a feature vector based on the acoustic feature parameters comprise: calculating the pure time-frequency domain acoustic signal to extract multi-dimensional acoustic feature parameters, wherein the acoustic feature parameters include equivalent continuous sound pressure level, peak sound pressure level, spectral centroid, and signal bandwidth; combining the multi-dimensional acoustic feature parameters to construct a feature vector for model matching.

[0010] Preferably, the noise reduction filtering processing comprises noise threshold judgment and A-weighted filtering, and the analysis frequency band is limited within the characteristic vibration frequency range of the vibration motor.

[0011] Preferably, the construction of the pre-trained vibration performance evaluation model comprises: collecting acoustic feature data of a plurality of standard motor samples with known vibration performance levels to construct a training set; training a neural network model using the training set to establish a mapping relationship from an acoustic feature space to a vibration performance level, and obtaining a pre-trained vibration performance evaluation model.

[0012] Preferably, the pre-trained vibration performance evaluation model is a classification model capable of distinguishing different motor types and different vibration intensity levels, and the quantitative evaluation result includes motor type and vibration intensity.

[0013] Preferably, after outputting the quantitative evaluation result of the vibration motor performance, the method further comprises mapping the quantitative evaluation result to a performance rating and generating a visual chart, wherein the visual chart includes a sound pressure level-time curve and a frequency spectrum analysis graph.

[0014] According to a second aspect of the present application, there is provided an apparatus motor vibration performance acoustic quantitative evaluation device, comprising: a signal acquisition module, starting a vibration motor of a device to be measured, and acquiring an original audio signal generated by the vibration motor in operation; a preprocessing module, preprocessing the original audio signal to obtain a preprocessed acoustic signal; a feature extraction module, extracting acoustic feature parameters based on the preprocessed acoustic signal, and constructing a feature vector based on the acoustic feature parameters; an evaluation module, inputting the feature vector into a pre-trained vibration performance evaluation model, and outputting a quantitative evaluation result of the vibration motor performance by comparing the feature vector with preset standard feature data.

[0015] According to a third aspect of the present application, there is provided an electronic device, comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method of acoustic quantitative evaluation of the vibration performance of the device motor according to the first aspect.

[0016] According to a fourth aspect of the present application, there is provided a non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the method of acoustic quantitative evaluation of the vibration performance of the device motor according to the first aspect.

[0017] According to a fifth aspect of the present application, there is provided a computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method of acoustic quantitative evaluation of the vibration performance of the device motor according to the first aspect.

[0018] In the embodiment provided by the application, by collecting the acoustic signal of motor vibration and extracting quantifiable feature parameters, an accurate mapping between vibration performance and objective data is established. The pre-trained machine learning model is used for automatic matching and grading, and a consistent quantitative result is output, which eliminates human differences and provides reliable and unified standards for product quality control; without relying on expensive or complex special hardware devices such as laser vibration meters and externally attached acceleration sensors, the test cost and deployment threshold are significantly reduced; the whole process including motor driving, signal collection, feature analysis, model matching, result judgment and report generation can be fully automated through software API (application programming interface) and script, which improves the test efficiency and automation level; by collecting the mechanical noise radiated by the motor during operation, the acoustic signal actually contains the information of the whole conduction path from the vibration energy inside the motor, through the machine body structure and finally causing the shell vibration, which indirectly reflects the overall performance of the vibration in the real user's hand-held state, makes up for the limitation of the traditional patch sensor that only measures the vibration of local point, and makes the performance evaluation closer to the tactile perception experience of the end user.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings are used to better understand the present application, and do not constitute a limitation on the present application. Among them: Figure 1 A flowchart of an acoustic quantitative evaluation method of a device motor vibration performance provided by the related art; Figure 2 A flowchart of an acoustic quantitative evaluation method of an Android device motor vibration performance provided by the embodiment of the application; Figure 3 A structural diagram of an acoustic quantitative evaluation device of a device motor vibration performance provided by the embodiment of the application. DETAILED DESCRIPTION

[0021] The exemplary embodiments of the application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the application to help understanding, and should be considered only as exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the application. Similarly, for the sake of clarity and conciseness, the description below omits the description of well-known functions and structures.

[0022] A method and device for acoustic quantitative evaluation of motor vibration performance of a device are described below with reference to the accompanying drawings.

[0023] Figure 1 A flowchart of an acoustic quantitative evaluation method of a device motor vibration performance provided by an embodiment of the present application is shown in FIG. 1. As shown in the figure, the method comprises the following steps: Figure 1 Step S101: Start the vibration motor of the device to be tested, and obtain the original audio signal generated by the vibration motor in operation; Step S102: Preprocess the original audio signal to obtain a preprocessed acoustic signal; Step S103: Extract acoustic feature parameters based on the preprocessed acoustic signal, and construct a feature vector based on the acoustic feature parameters; Step S104: Input the feature vector into a pre-trained vibration performance evaluation model, compare the feature vector with preset standard feature data, and output a quantitative evaluation result of the vibration motor performance.

[0024] Based on the above embodiment, step S101 is described in detail as follows: In one embodiment, the motor is directly controlled to start through an Android underlying API or a HAL layer interface, and is ensured to operate under a standard operating voltage and duty cycle to obtain reproducible acoustic data; The AudioRecord (a core class of the Android system) is used to control the device microphone, and recording parameters such as a sampling rate (48 kHz is recommended to cover the frequency that can be sensed by human ears), a channel (single channel), and an encoding format (such as ENCODING_PCM_16BIT) are set. The start of recording should be synchronized with the motor vibration, and a motor driving completion signal can be monitored through a callback or a start delay can be set to achieve this.

[0025] The thread or HandlerThread (a tool class of the Android system) is used to manage the audio acquisition thread to avoid blocking of the main thread or audio frame loss during recording.

[0026] Based on the above embodiment, step S102 is described in detail as follows: In one embodiment, the original audio signal is subjected to time-frequency transformation and noise reduction filtering processing to obtain a pure time-frequency domain acoustic signal; Noise threshold judgment and noise reduction processing (such as using a Wiener filter or a moving average filter) are performed to remove the influence of background noise and improve the purity of the sampling data; The background noise baseline is extracted and normalized to ensure that the sound pressure level analysis is concentrated in the motor signal frequency band (such as 120 Hz-300 Hz for ERM (eccentric rotating motor) and 170 Hz-250 Hz for LRA (linear resonant motor)). ​

[0027] Specifically, the recorded PCM data (pulse code modulation data) is subjected to short-time Fourier transform (STFT) to extract time-frequency features, using an integrated FFT library (software function library of fast Fourier transform algorithm) in Android or a third party such as JTransform (computing library of Java platform).

[0028] Through noise reduction and filtering, the environmental noise and electrical interference are effectively stripped off, ensuring that the analysis object is the acoustic signal purely generated by motor vibration, avoiding pollution of irrelevant factors on the evaluation result. Through band limitation and normalization processing, the analysis is focused on the core frequency band that best represents the motor working characteristics, enhancing the correlation between the signal and the performance to be evaluated. The audio signals from different devices and different environments are in a comparable baseline state before analysis, ensuring the consistency and robustness of the evaluation algorithm.

[0029] Based on the above embodiment, step S103 is described in detail: In one embodiment, the pure time-frequency domain acoustic signal is calculated to extract multi-dimensional acoustic feature parameters, including equivalent continuous sound pressure level, peak sound pressure level, spectral centroid and signal bandwidth. The multi-dimensional acoustic feature parameters are combined to construct a feature vector for model matching.

[0030] Specifically, A-weighting filter is used to adjust the human ear sensitivity of the signal, and the effective sound pressure level (SPL) of the unit time window is calculated, and the calculation formula is:

[0031] wherein, is the effective value of the signal, is the reference sound pressure value; Peak SPL, Leq, spectral centroid, bandwidth, modulation frequency and other indicators are extracted to reflect the motor vibration intensity, frequency consistency and structural stability. A feature vector is constructed for subsequent model training.

[0032] The audio waveform is condensed into a few core feature values with clear physical meaning, greatly reducing the computational burden of the subsequent model and improving efficiency. Each parameter extracted corresponds to a specific dimension of motor performance, so that the abstract vibration experience is translated into a quantitative language that machines can understand and process.

[0033] Based on the above embodiment, step S104 is described in detail: In one embodiment, the pre-trained vibration performance evaluation model is constructed by: Collecting acoustic characteristic data of a plurality of standard motor samples with known vibration performance levels to construct a training set; Training a neural network model using the training set to establish a mapping relationship from the acoustic characteristic space to the vibration performance level, and obtaining a pre-trained vibration performance evaluation model; The pre-trained vibration performance evaluation model is a classification model capable of distinguishing different motor types and different vibration intensity levels, and the quantitative evaluation result includes the motor type and the vibration intensity.

[0034] Specifically, the system scores the vibration performance of the measured motor according to the acoustic characteristics and the matching model, and maps the score value to the "A-E" level, or distributes the intensity to three levels of "weak, medium, and strong".

[0035] Further comprising: outputting the quantitative evaluation result of the vibration motor performance, including mapping the quantitative evaluation result to a performance rating and generating a visualization chart, the visualization chart including a sound pressure level curve over time and a frequency spectrum analysis graph.

[0036] Specifically, the chart library (such as MPAndroidChart) is used to generate sound pressure curve graphs, spectrum graphs, and gear radar graphs in real time, so that the tester can intuitively understand the motor characteristics.

[0037] The result export function is provided to support outputting the score report in JSON (data exchange format), CSV (table data storage format), PDF (fixed layout document format), etc., for production line archiving or third-party platform uploading.

[0038] Through the level, gear, and visualization chart, the complex algorithm output is translated into a performance conclusion and intuitive image that can be instantly understood by humans, facilitating the tester to quickly judge and make decisions.

[0039] This embodiment briefly describes the system management and user interface, as follows: An operation interface based on the MVVM (Model-View-ViewModel, software architecture design pattern) architecture is provided, allowing users to select motor types, set recording times, play back recordings, view historical records, etc.

[0040] Device permission dynamic application (such as RECORD_AUDIO recording permission, VIBRATE vibration permission) is supported, and permission verification is performed to ensure that the system runs legally and in compliance.

[0041] A developer mode is provided to adjust recording parameters, motor driving methods, and acoustic model debugging parameters, meeting the diversified testing needs in the research and development stage.

[0042] The embodiment provides an acoustic quantitative evaluation method of motor vibration performance of a device, and establishes an accurate mapping between vibration performance and objective data by collecting acoustic signals of motor vibration and extracting quantifiable characteristic parameters. A pre-trained machine learning model is used for automatic matching and grading, and a consistent quantitative result is output, so as to eliminate human differences and provide reliable and unified standards for product quality control. The method does not need to rely on expensive or complex special hardware devices such as a laser vibration meter and an externally attached acceleration sensor, significantly reduces the test cost and deployment threshold, can be fully automatically executed through a software API (application programming interface) and a script, including a whole process of motor driving, signal collection, characteristic analysis, model matching, result judgment and report generation, and improves the test efficiency and the automation level. The mechanical noise radiated by the motor during operation is collected, the acoustic signal actually contains information of a whole conduction path from vibration energy in the motor to the shell vibration caused by the structure of the motor, indirectly reflects the overall performance of the vibration in the real user's handheld state, makes up for the limitation of the traditional patch sensor that only measures local point vibration, and makes the performance evaluation closer to the tactile perception experience of the end user.

[0043] Based on the above embodiment, the embodiment describes an acoustic quantitative evaluation implementation process of motor vibration performance of an Android device, as shown in Figure 2 , and the implementation process is as follows: System initialization and permission preparation: A test terminal (an Android smart phone) starts a built-in "motor test application". After the application is started, the application first dynamically applies for RECORD_AUDIO (recording) and VIBRATE (vibration) permissions from the operating system. After the user grants the permission, the application loads a preset acoustic evaluation model and initializes audio acquisition parameters.

[0044] Triggering standardized vibration and synchronous recording: The LRA (linear resonant motor) of the control device is started at a "standard intensity level" to generate a vibration with a certain duration. At the same time when the vibration control instruction is sent, the application starts recording immediately through an AudioRecord class instance, so as to ensure that the recording window completely covers the whole process of the motor from starting, steady vibration to stopping. This process is managed through a background thread to avoid interface lag.

[0045] Preprocessing of sound signals: After the recording is finished, the application obtains a raw PCM (Pulse Code Modulation) audio data. First, a high-pass filter is applied to remove the low-frequency background noise in the environment. Then, the filtered audio data is subjected to a short-time Fourier transform, converting it from the time domain to the time-frequency domain, and obtaining a spectrogram. Then, according to the known characteristic frequency band of this type of LRA (Linear Resonant Actuator), the target analysis area is framed on the spectrogram, and the average energy outside the area is calculated as the noise baseline. Finally, the effect of the noise baseline is subtracted from the original signal to obtain the enhanced target acoustic signal dominated by the motor vibration sound.

[0046] Extraction of acoustic feature parameters: The pre-processed target acoustic signal is subjected to A-weighted filtering to simulate the human ear's hearing characteristics. On the A-weighted signal, the following key feature parameters are calculated: Equivalent continuous sound pressure level: reflects the average sound intensity during the entire vibration process. Peak sound pressure level: reflects the sound intensity corresponding to the strongest moment of vibration. Spectral centroid: reflects the concentration position of vibration sound energy in the frequency dimension. One-third octave bandwidth: reflects the frequency distribution range of the main vibration energy.

[0047] The above four parameter values are combined to form a four-dimensional acoustic feature vector.

[0048] Model-based performance matching and evaluation: The acoustic feature vector obtained in the previous step is input into the loaded vibration performance evaluation model. This model is a multi-classification support vector machine model that has been pre-trained through a large number of good-quality motor samples of the same type.

[0049] The model first confirms the type of the measured motor and the pre-set LRA model based on the characteristics of the "spectral centroid" and "bandwidth" in the feature vector.

[0050] Subsequently, the core classifier of the model calculates the distance or similarity between the entire feature vector and the standard features of each pre-defined performance level (such as "Excellent-A level", "Good-B level", "Qualified-C level", "Unqualified") based on the entire feature vector.

[0051] The model finally outputs a judgment result, such as "B level". At the same time, the model may also output a confidence score indicating the reliability of this judgment.

[0052] Result visualization and report generation: The application interface presents the evaluation results in an intuitive format, displaying a prominent grade label in the center of the screen, such as "Grade B (Good)". Below, two curves are plotted: one is the waveform of sound pressure level changing over time, and the other is a spectral analysis graph of the main vibration stages. Testers can intuitively judge the stability and spectral purity of the vibration signal through these graphs. Testers can click the "Save Report" button. The application automatically generates a structured JSON report from the raw test data, feature vectors, model judgment results, timestamps, and equipment serial number, storing it locally on the device or uploading it to the production line server for quality traceability and analysis.

[0053] Exception handling and process termination: If the confidence level of the evaluation model output is lower than the preset threshold, or if the recording is interrupted, the application will determine the test as "invalid" and display the message "Test failed, please try again" on the interface. The tester can trigger a retest. If the test is successful and the report is saved, the evaluation process ends, and the device can proceed to the next test or be taken offline.

[0054] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0055] According to embodiments of the present invention, the present invention also provides an acoustic quantitative evaluation device for the vibration performance of a device motor. For example, Figure 3 This is a schematic diagram of the structure of an acoustic quantitative evaluation device for the vibration performance of a device motor provided in an embodiment of the present invention. The acoustic quantitative evaluation device 300 for the vibration performance of the device motor includes: The signal acquisition module 310 starts the vibration motor of the device under test and acquires the original audio signal generated by the vibration motor. The preprocessing module 320 preprocesses the original audio signal to obtain a preprocessed acoustic signal; Feature extraction module 330 extracts acoustic feature parameters based on the preprocessed acoustic signal and constructs a feature vector based on the acoustic feature parameters; The evaluation module 340 inputs the feature vector into a pre-trained vibration performance evaluation model, and outputs a quantitative evaluation result of the vibration motor performance by comparing the feature vector with preset standard feature data.

[0056] It should be noted that the description of the features in the embodiment corresponding to the acoustic quantitative evaluation device for the vibration performance of the equipment motor can be found in the relevant description of the embodiment corresponding to the acoustic quantitative evaluation method for the vibration performance of the equipment motor, and will not be repeated here.

[0057] Embodiments of the present invention also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in the embodiments of the acoustic quantitative evaluation method for the vibration performance of a device motor described above.

[0058] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program configured to execute the steps in the embodiments of the acoustic quantitative evaluation method for the vibration performance of a device motor described above when the computer program is run.

[0059] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0060] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in the embodiments of the acoustic quantitative evaluation method for the vibration performance of a device motor described above.

[0061] Embodiments of the present invention also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the embodiments of the acoustic quantitative evaluation method for the vibration performance of a device motor described above.

[0062] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0063] The acoustic quantitative evaluation method for the vibration performance of a device motor provided by this invention has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. It should be noted that those skilled in the art can make several improvements and modifications to this invention without departing from the principles of this invention, and these improvements and modifications also fall within the protection scope of the claims of this invention.

Claims

1. An acoustic quantitative evaluation method for the vibration performance of a device motor, characterized in that, include: Start the vibration motor of the device under test and acquire the original audio signal generated by the vibration motor. The original audio signal is preprocessed to obtain a preprocessed acoustic signal; Based on the preprocessed acoustic signal, acoustic feature parameters are extracted, and feature vectors are constructed based on the acoustic feature parameters; The feature vector is input into a pre-trained vibration performance evaluation model. By comparing the feature vector with preset standard feature data, a quantitative evaluation result of the vibration motor performance is output.

2. The acoustic quantitative evaluation method for the vibration performance of a device motor according to claim 1, characterized in that, The preprocessing of the original audio signal to obtain the preprocessed acoustic signal includes: performing time-frequency transformation and noise reduction filtering on the original audio signal to obtain a clean time-frequency domain acoustic signal.

3. The acoustic quantitative evaluation method for the vibration performance of a device motor according to claim 2, characterized in that, The step of extracting acoustic feature parameters based on the preprocessed acoustic signal and constructing a feature vector based on the acoustic feature parameters includes: The pure time-frequency domain acoustic signal is calculated to extract multi-dimensional acoustic feature parameters, including equivalent continuous sound pressure level, peak sound pressure level, spectral centroid, and signal bandwidth. The multi-dimensional acoustic feature parameters are combined to construct a feature vector for model matching.

4. The acoustic quantitative evaluation method for the vibration performance of a device motor according to claim 2, characterized in that, The noise reduction filtering process includes: using noise threshold judgment and A-weighted filtering, and limiting the analysis frequency band to the characteristic vibration frequency range of the vibration motor.

5. The acoustic quantitative evaluation method for the vibration performance of a device motor according to claim 1, characterized in that, The construction of the pre-trained vibration performance evaluation model includes: Acoustic feature data of multiple standard motor samples with known vibration performance levels were collected to construct a training set; The neural network model is trained using the training set to establish a mapping relationship from acoustic feature space to vibration performance level, thereby obtaining a pre-trained vibration performance evaluation model.

6. The acoustic quantitative evaluation method for the vibration performance of a device motor according to claim 5, characterized in that, The pre-trained vibration performance evaluation model is a classification model that can distinguish between different motor types and different vibration intensity levels. The quantitative evaluation results include motor type and vibration intensity.

7. The acoustic quantitative evaluation method for the vibration performance of a device motor according to claim 1, characterized in that, The output of the quantitative evaluation result of the vibration motor performance includes: mapping the quantitative evaluation result to a performance rating and generating a visualization chart, the visualization chart including a sound pressure level change curve over time and a spectrum analysis graph.

8. An acoustic quantitative evaluation device for the vibration performance of a device motor, characterized in that, include: The signal acquisition module starts the vibration motor of the device under test and acquires the original audio signal generated by the vibration motor. The preprocessing module preprocesses the original audio signal to obtain a preprocessed acoustic signal; The feature extraction module extracts acoustic feature parameters based on the preprocessed acoustic signal and constructs a feature vector based on the acoustic feature parameters. The evaluation module inputs the feature vector into a pre-trained vibration performance evaluation model, and outputs a quantitative evaluation result of the vibration motor performance by comparing the feature vector with preset standard feature data.

9. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the acoustic quantitative evaluation method for the vibration performance of the device motor as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to execute the acoustic quantitative evaluation method for the vibration performance of the device motor as described in any one of claims 1-7.

Images