Automobile acceleration noise evaluation method and device and storage medium
By using MFCC features and tone mutation evaluation in automotive acceleration noise evaluation, the problem of failure to match the auditory characteristics of the human ear in the prior art is solved, and a more accurate and detailed evaluation of automotive acceleration noise is achieved to meet high-quality requirements.
Patent Information
- Application Number
- CN202411042389.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-07-31
AI Technical Summary
The existing automotive acceleration noise evaluation methods fail to fully consider the auditory characteristics of the human ear, resulting in the mismatch between the evaluation results and the user's subjective feelings and insufficient accuracy.
The MelFrequency Cepstrum Coefficient (MelFrequency Cepstrum Coefficient, MFCC) was used to evaluate the vehicle acceleration noise. By obtaining the noise signal under the engine speed range, the MFCC characteristics were calculated in segments, and the results of the vehicle acceleration noise evaluation were determined based on the results of tone mutation evaluation.
It improves the accuracy and comprehensiveness of the evaluation of automobile acceleration noise, meets people's requirements for high-quality acceleration noise, and matches the evaluation results with the auditory characteristics of the human ear.
Smart Images

Figure CN118913714A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of automobile performance analysis, and in particular to an automobile acceleration noise evaluation method, equipment and storage medium. Background Art
[0002] The objective evaluation method for automobile acceleration noise is generally: after measuring the noise inside the car and the engine speed signal, the curve of the change of the sound pressure level inside the car with the engine speed is obtained through data processing. For example, the Chinese invention patent with the publication number CN117109938A discloses a method and device for evaluating the quality of vehicle acceleration noise. The steps of the method are: generating an acceleration noise waterfall diagram according to the total value of the noise inside the car when the vehicle accelerates; extracting noise energy according to the order of the noise in the acceleration noise waterfall diagram according to the preset order bandwidth to obtain the powertrain acceleration noise energy; determining the background noise energy when the vehicle accelerates according to the energy of the total noise value and the powertrain acceleration noise energy; evaluating the acceleration noise quality of the vehicle according to the relationship between the powertrain acceleration noise energy and the background noise energy and the corresponding acceleration noise evaluation index. It realizes the comprehensive evaluation of the acceleration noise quality of the vehicle from the two aspects of the vehicle powertrain acceleration noise and the background noise, thereby helping the vehicle to accurately locate and provide guidance for the development of automobile noise quality. Summary of the invention
[0003] In view of the defects existing in the prior art, the technical problem solved by the present invention is: how to evaluate the automobile acceleration noise based on MFCC (Mel Frequency Cepstrum Coefficient) so that the evaluation result of the automobile acceleration noise matches the auditory characteristics of the human ear, that is, conforms to the subjective feelings of the user, thereby improving the evaluation accuracy.
[0004] To achieve the above objectives, in a first aspect, an embodiment of the present application provides a method for evaluating automobile acceleration noise, the method comprising the following steps:
[0005] Acquire a noise signal set within an engine speed range, and segment the noise signal set to form a plurality of noise signal data segments;
[0006] Determine the MFCC features of each noise signal data segment, and determine the MFCC evaluation result according to each MFCC feature and its corresponding engine speed;
[0007] The timbre mutation evaluation result is determined based on the similarity of the MFCC features of adjacent segments;
[0008] The evaluation results of automobile acceleration noise are determined based on the MFCC evaluation results and the timbre mutation evaluation results.
[0009] In combination with the first aspect, in one implementation, the process of obtaining a noise signal set within an engine speed range includes:
[0010] respectively acquiring an engine lower speed limit closest to a lower speed limit of the engine speed range and an engine upper speed limit closest to an upper speed limit of the engine speed range;
[0011] forming an engine speed set according to an upper engine speed limit, a lower engine speed limit, and other engine speeds between the upper engine speed limit and the lower engine speed limit;
[0012] The noise signals corresponding to all engine speeds in the engine speed set are taken as the noise signal set.
[0013] In combination with the first aspect, in one implementation, the calculation formula for the segmentation time length T for segmenting the noise signal set is:
[0014] T=N*Δt, N represents the number of sampling points of the noise signal in each noise signal set after segmentation, Δt represents the noise sampling interval, Δt=1 / fs, fs is the noise sampling frequency.
[0015] In combination with the first aspect, in one implementation, the process of segmenting the noise signal set to form a plurality of segments of noise signal data includes:
[0016] Determine the total duration according to the minimum sampling time and the maximum sampling time in the noise signal set, and determine the time range of each segment according to the total duration and the segment duration;
[0017] The noise signal at the sampling moment belonging to each time range is taken as the noise signal data of that section.
[0018] In combination with the first aspect, in one implementation, the process of determining the MFCC features of each segment of noise signal data and determining the MFCC evaluation result according to each MFCC feature and its corresponding engine speed includes:
[0019] Determine the MFCC features of each segment of noise signal data, and calculate the MFCC amplitude of each MFCC feature vector;
[0020] Based on each MFCC amplitude and its corresponding engine speed, an evaluation curve of MFCC amplitude changing with engine speed is formed;
[0021] The MFCC evaluation result is determined according to the evaluation curve.
[0022] In combination with the first aspect, in one implementation, the process of forming an evaluation curve of MFCC amplitude changing with engine speed according to each MFCC amplitude and its corresponding engine speed includes:
[0023] The engine speed and MFCC amplitude are used as the horizontal and vertical coordinates respectively;
[0024] Determine the coordinate point according to each MFCC amplitude and its corresponding engine speed;
[0025] Connect all coordinate points to obtain the evaluation curve of MFCC amplitude changing with engine speed;
[0026] The process of determining the MFCC evaluation result according to the evaluation curve includes: generating a target line corresponding to the vehicle acceleration noise target requirement, judging whether the evaluation curve intersects with the target line, and if so, determining that the MFCC evaluation result is not passed, otherwise determining that the MFCC evaluation result is passed.
[0027] In combination with the first aspect, in one embodiment, the process of determining the timbre mutation evaluation result based on the similarity of adjacent MFCC features includes: calculating the cosine similarity of the MFCC feature vectors of adjacent segmented noise signal data; judging whether the number of times the cosine similarity is less than a reference threshold exceeds a specified value, and if so, determining that the timbre mutation evaluation result fails, otherwise determining that the timbre mutation evaluation result passes.
[0028] In combination with the first aspect, in one embodiment, the process of determining the automobile acceleration noise evaluation result based on the MFCC evaluation result and the timbre mutation evaluation result includes: judging whether both the MFCC evaluation result and the timbre mutation evaluation result are passed; if so, determining that the automobile acceleration noise evaluation result is passed; otherwise, determining that the automobile acceleration noise evaluation result is failed.
[0029] In a second aspect, an embodiment of the present application provides a vehicle acceleration noise evaluation device, the vehicle acceleration noise evaluation device comprising a processor, a memory, and a vehicle acceleration noise evaluation program stored in the memory and executable by the processor, wherein when the vehicle acceleration noise evaluation program is executed by the processor, the method provided in the first aspect is implemented.
[0030] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, characterized in that a vehicle acceleration noise evaluation program is stored on the computer-readable storage medium, wherein when the vehicle acceleration noise evaluation program is executed, the method provided in the first aspect is implemented.
[0031] Compared with the prior art, the advantages of the present invention are:
[0032] The present invention adopts MFCC features to evaluate automobile acceleration noise, fully considering the auditory characteristics of the human ear; at the same time, the present invention also determines the MFCC evaluation result of the automobile acceleration noise according to each MFCC feature and its corresponding engine speed; and determines the timbre mutation evaluation result according to the similarity of adjacent segmented MFCC features, to jointly determine the automobile acceleration noise evaluation result, the evaluation result obtained thereby is relatively accurate, comprehensive and detailed, and can meet people's requirements for high quality of acceleration noise. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0034] Figure 1 is a schematic diagram of an evaluation curve of MFCC amplitude changing with engine speed in an embodiment of the present invention;
[0035] Figure 2 is a schematic diagram of comparison between a target line and an evaluation curve in an embodiment of the present invention;
[0036] Figure 3 is a schematic diagram for comparing a segmented target line and an evaluation curve in an embodiment of the present invention;
[0037] Figure 4 It is a linearity schematic diagram showing that the difference between the upper limit and the lower limit of the evaluation curve does not change with the engine speed in the embodiment of the present invention;
[0038] Figure 5 It is a schematic diagram of the linearity of the difference between the upper limit and the lower limit of the evaluation curve in the embodiment of the present invention as the engine speed changes;
[0039] Figure 6 Schematic diagram of the process of the vehicle acceleration noise evaluation method in an embodiment of the present invention;
[0040] Figure 7 This is a schematic diagram of the hardware structure of the automobile acceleration noise evaluation device involved in the embodiment of the present application. DETAILED DESCRIPTION
[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are 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 ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0042] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.
[0043] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.
[0044] First, the research and development process of the present invention is introduced.
[0045] The inventors found through subjective feelings and market research that the acceleration noise evaluation results in the prior art may be different from the user's feelings. For example, the acceleration noise evaluation result of car A meets the requirements, but when the user uses car A, he or she may perceive uncomfortable (such as "harsh") noise.
[0046] The inventors further analyzed and concluded that the reason for the above situation is that the acceleration noise evaluation method in the prior art does not fully consider the auditory characteristics of the human ear.
[0047] To this end, the inventors conducted research and development and obtained that the results of accelerated noise evaluation based on MFCC can adapt to the auditory characteristics of the human ear, specifically: Mel is the unit of subjective pitch, Hz (Hertz) is the unit of objective pitch; Mel frequency is proposed based on the auditory characteristics of the human ear, and it has a nonlinear relationship with Hz (Hertz) frequency. Mel frequency cepstral coefficients (MFCC) are cepstral features calculated using this relationship between them.
[0048] On this basis, in the first aspect, an embodiment of the present application provides a method for evaluating automobile acceleration noise, the steps of which include: obtaining a noise signal set under an engine speed range that needs to be evaluated, and segmenting the noise signal set to form a plurality of noise signal data segments (data blocks); determining the MFCC features of each noise signal data segment, and determining the MFCC evaluation result according to each MFCC feature and its corresponding engine speed; determining the timbre mutation evaluation result according to the similarity of adjacent segmented MFCC features; and determining the automobile acceleration noise evaluation result according to the MFCC evaluation result and the timbre mutation evaluation result.
[0049] It can be seen that the present invention uses MFCC features to evaluate automobile acceleration noise, fully considering the auditory characteristics of the human ear; at the same time, the present invention also determines the MFCC evaluation result of automobile acceleration noise according to each MFCC feature and its corresponding engine speed; and determines the timbre mutation evaluation result according to the similarity of adjacent segmented MFCC features, to jointly determine the automobile acceleration noise evaluation result. The evaluation result obtained in this way is relatively accurate, comprehensive and detailed, and can meet people's requirements for high quality of acceleration noise.
[0050] In one embodiment, the engine speed range to be evaluated is explained as: evaluating the engine speed range corresponding to the acceleration of the vehicle speed from 0 to a specified speed, for example, 1000-5000 rpm, where rpm is revolutions per minute.
[0051] The process of obtaining a set of noise signals within the engine speed range that needs to be evaluated includes: respectively obtaining (sampling) the engine lower limit speed that is closest to the lower limit value of the engine speed range, and the engine upper limit speed that is closest to the upper limit value of the engine speed range; forming an engine speed set (in array form during execution, i.e., an engine speed sequence) based on the engine upper limit speed, the engine lower limit speed, and other engine speeds between the engine upper limit speed and the engine lower limit speed.
[0052] The noise signal closest to the sampling time of the engine speed (the noise signal is obtained by sampling, and the sampling method and sampling frequency of the engine speed may be the same or different, and there is no restriction here) is defined as the noise signal corresponding to the engine speed; the noise signals corresponding to all engine speeds in the engine speed set are taken as the noise signal set (in array mode during execution, that is, noise signal sequence).
[0053] Furthermore, the reason why the above-mentioned closest values are not the same is that the engine speed range is a theoretical value, and the engine speed is obtained by actual sampling. It may happen that the upper limit or lower limit is not sampled. Therefore:
[0054] The above-mentioned process of determining the upper and lower speed limits of the engine includes:
[0055] When the engine speed sampled is the same as the upper limit value or the lower limit value, the engine speed is used as the upper limit speed or the lower limit speed;
[0056] When no engine speed that is the same as the upper limit or the lower limit is sampled, the engine speed with the smallest absolute difference from the upper limit or the lower limit is selected as the upper limit or the lower limit;
[0057] When no engine speed that is the same as the upper limit value or the lower limit value is sampled, and there are two or more engine speeds whose absolute difference from the upper limit value or the lower limit value is the smallest and whose difference is the same: for the lower limit value, the smallest engine speed is selected as the lower limit speed, and for the upper limit value, the largest engine speed is selected as the upper limit speed.
[0058] The reason for performing the above operation is that the sources of the noise signal and the engine speed signal are different. The in-car noise signal comes from the microphone, and the engine speed signal comes from the speed sensor or the vehicle CAN signal. Therefore, the sampling frequencies of these two signals are generally different. The sampling frequency of the in-car noise is recorded as fs, which is generally 40000Hz or 20000Hz; the sampling frequency of the engine speed signal is lower, generally a few hundred hertz, such as 256Hz, that is, the two types of signals are not synchronized. Therefore, it is necessary to combine the engine speed range and perform segmented processing on the noise signal set under the engine speed range.
[0059] On this basis, the calculation formula for the segmentation time length T of the noise signal set in the above method is: T = 1 / Δf = N / fs = N*Δt, where N represents the number of sampling points of the noise signal in each noise signal set after segmentation, N is an integer, Δt represents the noise sampling time interval, Δf represents the frequency resolution of the spectrum analysis, fs is the noise sampling frequency, and Δt = 1 / fs.
[0060] Furthermore, the process of segmenting the noise signal set in the above method to form several segments of noise signal data includes: determining the total duration according to the minimum sampling time and the maximum sampling time in the noise signal set, and determining the time range of each segment according to the total duration and the above segment duration; for example, the acquisition segment duration is T, and the minimum sampling time is t(1), then the time range of the first segment of noise signal is [t(1), t(1)+T); the noise signal whose acquisition time belongs to each time range is used as the noise signal data of that segment.
[0061] It should be noted that the sampling time in the previous paragraph can be the sampling time of the engine speed corresponding to the noise signal, or it can be the sampling time of the noise signal itself. Regardless of which one, this solution can be implemented without limitation here.
[0062] In one embodiment, the process of determining the MFCC features of each noise signal data segment in the above method and determining the MFCC evaluation result according to each MFCC feature and its corresponding engine speed includes: determining the MFCC features of each noise signal data segment (the calculation of the MFCC features is a known technology), and calculating the MFCC amplitude amp of each MFCC feature vector, and the calculation formula is: Where M represents the number of elements in the MFCC feature vector, and Ci represents the elements in the MFCC feature vector.
[0063] According to each MFCC amplitude and its corresponding engine speed, an evaluation curve of MFCC amplitude changing with engine speed is formed; and the MFCC evaluation result is determined according to the evaluation curve.
[0064] It should be noted that the above-mentioned MFCC features or MFCC amplitudes and their corresponding engine speeds are the engine speeds corresponding to the noise signals specified in the noise signal data to which the MFCC features or MFCC amplitudes belong, and the specified noise signals are the minimum or maximum noise signals at the time of acquisition.
[0065] Specifically, the process of forming an evaluation curve of MFCC amplitude changing with engine speed according to each MFCC amplitude and its corresponding engine speed includes: Figure 1 As shown, a plane coordinate system is established, with the engine speed and MFCC amplitude as the horizontal and vertical coordinates respectively. The coordinate points are determined according to each MFCC amplitude and its corresponding engine speed, and all coordinate points are connected to obtain an evaluation curve of the MFCC amplitude changing with the engine speed. Generally speaking, the higher the speed, the larger the MFCC amplitude. Therefore, the evaluation curve is a curve that rises from the lower left corner to the upper right corner.
[0066] Further, the above process of determining the MFCC evaluation result according to the evaluation curve includes: Figure 2 As shown, a target line corresponding to the target requirement of vehicle acceleration noise is generated, and it is determined whether the evaluation curve intersects with the target line. If so, the MFCC evaluation result is determined to be failed, otherwise, the MFCC evaluation result is determined to be passed.
[0067] It should be noted that see Figure 3 As shown, when judging by the above target line, a segmented target line can be set, that is, different requirements are placed on the slope of the target line in different speed sections.
[0068] At the same time, the above process of determining the MFCC evaluation result according to the evaluation curve also includes: setting the linearity requirement of the evaluation curve, and when the evaluation curve does not intersect the target line and the linearity meets the requirement, determining that the MFCC evaluation result is passed.
[0069] The reason for doing so is that setting linearity requirements can ensure that the evaluation curve has a smaller tolerance at each speed, making the evaluation curve more linear and making the in-vehicle noise feel smoother and more silky.
[0070] The specific way to set the linearity requirement is: Figure 4As shown, according to the principle of least squares method, a slope is fitted to the level vsrpm curve, which is called the middle line; in the present invention, the linearity tolerance is used to evaluate the linearity, and the calculation formula of the linearity tolerance is: Among them, a is the maximum distance between the evaluation curve and the middle line along the ordinate, and current_val is the ordinate of the point on the middle line. From the calculation formula, we know that the smaller the linearity tolerance, the better the linearity; the larger the linearity tolerance, the worse the linearity.
[0071] Furthermore, this embodiment provides two methods for setting linearity requirements:
[0072] (1) When current_val is a constant (the ordinate of a point on the middle line), after the linearity tolerance linearity is determined, a is a fixed value. Draw a slash line on both sides of the middle line with a longitudinal distance a from the middle line. Both slash lines are parallel to the middle line, thereby determining the upper and lower limits of the evaluation curve, such as Figure 4 When the evaluation curve of the vehicle is between these two oblique lines, the linearity meets the requirement; otherwise, it does not meet the requirement.
[0073] (2) When current_val is not a constant and is the ordinate of the current point on the middle line, after the linearity tolerance linearity is determined, a changes with the change of the current point on the middle line. Draw an oblique line on both sides of the middle line, and the longitudinal distance between the point on the oblique line and the corresponding point on the middle line is a. The two oblique lines are not parallel to the middle line, but are trumpet-shaped; thus, the upper and lower limits of the evaluation curve are determined, such as Figure 5 As shown. When the evaluation curve is between these two oblique lines, the linearity meets the requirements; otherwise, it does not meet the requirements. Setting the linearity in the second way is realistic and reasonable. The higher the engine speed, the more uncontrollable factors there are (the greater the inertia force of the internal parts of the engine, the closer the excitation frequency is to the natural frequency of each component), the greater the noise inside the car, and the noise tolerance should be set more loosely.
[0074] In one embodiment, the process of determining the timbre mutation evaluation result according to the similarity of adjacent MFCC features in the above method includes: calculating the cosine similarity of the MFCC feature vectors of adjacent segmented noise signal data; judging whether the number of times the cosine similarity is less than a reference threshold exceeds a specified value, and if so, determining that the timbre mutation evaluation result fails, otherwise determining that the timbre mutation evaluation result passes.
[0075] It should be noted that: the above-mentioned reference threshold and specified value are set according to actual needs; the principle of the above-mentioned embodiment is: the range of cosine similarity is [-1,1], and the larger the value, the closer the two vectors are; the smaller the value, the less similar the two vectors are. Therefore, if the cosine similarity of adjacent segments is large, it is considered that the timbre has not changed much and the timbre has not changed suddenly; conversely, if the cosine similarity of adjacent segments is small, it is considered that the timbre has changed suddenly, which will give people an uncomfortable feeling and should be avoided. How to determine whether the cosine similarity is small, obviously, requires setting a threshold. Since the cosine similarity value range is [-1,1], naturally, this threshold can be set to 0. That is, if the cosine similarity ≤ 0, it is considered that the timbre has changed suddenly.
[0076] In one embodiment, the process of determining the vehicle acceleration noise evaluation result based on the MFCC evaluation result and the timbre mutation evaluation result in the above method includes: judging whether both the MFCC evaluation result and the timbre mutation evaluation result are passed, if so, determining that the vehicle acceleration noise evaluation result is passed, otherwise determining that the vehicle acceleration noise evaluation result is failed.
[0077] In summary, the present invention has a simple principle and a simple algorithm. The MFCC features used take into account the auditory characteristics of the human ear and are consistent with human subjective feelings. The target line, linearity, and timbre mutation are used to evaluate the acceleration noise, which is more accurate, comprehensive, and detailed, meeting people's high quality requirements for acceleration noise.
[0078] See below Figure 6 As shown, the method of the present invention is described through a specific embodiment.
[0079] S1: Obtain a set of noise signals within an engine speed range that needs to be evaluated.
[0080] The specific process of S1 includes:
[0081] S101: Setting the segment duration T of the noise signal, the calculation formula of T is as mentioned above.
[0082] S102: Set the engine speed range to be evaluated, specifically 1000-5000 rpm, where rpm is revolutions per minute. Set the lower limit of the speed range to low_rpm and the upper limit to up_rpm. Naturally, the set speed range must be within the actual measured speed range.
[0083] S103: Searching for speed signal: Find the speed data closest to low_rpm, recorded as rpm_array(1). rpm_array is the speed array, rpm_array(1) is its first element, and its time is recorded as t_temp; search for the sampling time closest to t_temp in the noise signal, recorded as t(1), t is the time array, and t(1) is its first element.
[0084] S2: Segment the noise signal set to form a number of noise signal data blocks.
[0085] The specific process of S2 includes:
[0086] Taking t(1) as the starting time of noise signal analysis and T as the segment duration, the first analysis data block is determined, and its time range is [t(1), t(1)+T); the speed corresponding to this data block is rpm_array(1).
[0087] Therefore, the time corresponding to the first analysis data block is t(1), and the corresponding speed is rpm_array(1). The starting time of the second analysis data block is t(2), t(2) = t(1) + T; then, the speed corresponding to the sampling time closest to t(2) is searched in the speed signal, recorded as rpm_array(2). Then the time range corresponding to the second analysis data block is [t(2), t(2) + T), the corresponding time is t(2), and the corresponding speed is rpm_array(2). ... The starting time of the jth analysis data block is t(j), t(j) = t(j-1) + T; then, the speed corresponding to the sampling time closest to t(j) is searched in the speed signal, recorded as rpm_array(j). Then the time range corresponding to the jth analysis data block is [t(j), t(j) + T), the corresponding time is t(j), and the corresponding speed is rpm_array(j). If rpm_array(j)>=up_rpm, S2 ends, and the arrays rpm_array and t are the speed sequence and time series; otherwise, continue this process.
[0088] As required, an overlap rate parameter can be set, that is, there are overlapping areas between adjacent analysis data blocks so that the noise signal can be fully utilized. The specific analysis process is similar to S2 and will not be repeated here.
[0089] S3: Determine the MFCC features of each noise signal data block, and calculate the MFCC amplitude of each MFCC feature vector. The calculation method is as described above.
[0090] S4: Based on each MFCC amplitude and its corresponding engine speed, an evaluation curve of the MFCC amplitude changing with the engine speed is formed; the MFCC evaluation result is determined based on the evaluation curve, and the determination method is: generate a target line corresponding to the vehicle acceleration noise target requirement, and determine whether the evaluation curve intersects with the target line. If so, the MFCC evaluation result is determined to be failed, otherwise, the MFCC evaluation result is determined to be passed.
[0091] S5: Determine the timbre mutation evaluation result according to the cosine similarity of the MFCC feature vectors of the noise signal data blocks of adjacent segments, specifically: determine whether the number of times the cosine similarity is less than the reference threshold exceeds a specified value, if so, determine that the timbre mutation evaluation result fails, otherwise determine that the timbre mutation evaluation result passes.
[0092] S6: Determine the evaluation result of the vehicle acceleration noise according to the MFCC evaluation result and the timbre mutation evaluation result, specifically: determine whether the MFCC evaluation result and the timbre mutation evaluation result are both passed, if so, determine that the evaluation result of the vehicle acceleration noise is passed, otherwise determine that the evaluation result of the vehicle acceleration noise is not passed.
[0093] In a second aspect, an embodiment of the present application provides a vehicle acceleration noise evaluation device, which may be a personal computer (PC), a laptop computer, a server, or other device with a data processing function.
[0094] Reference Figure 7 , Figure 7 The hardware structure diagram of the automobile device involved in the embodiment of the present application is shown in FIG. 1. In the embodiment of the present application, the automobile acceleration noise evaluation device may include a processor, a memory, a communication interface, and a communication bus.
[0095] The communication bus may be of any type and is used to interconnect the processor, the memory, and the communication interface.
[0096] The communication interface includes input / output (I / O) interface, physical interface and logical interface, etc., which are used to realize the interconnection of devices inside the vehicle acceleration noise evaluation device, and the interface used to realize the interconnection between the vehicle acceleration noise evaluation device and other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, an optical fiber interface, an ATM interface, etc.; the user device can be a display, a keyboard, etc.
[0097] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.
[0098] The processor may be a general-purpose processor, and the general-purpose processor may call the automobile acceleration noise evaluation program stored in the memory and execute the automobile acceleration noise evaluation method provided in the embodiment of the present application. For example, the general-purpose processor may be a central processing unit (CPU). The method executed when the automobile acceleration noise evaluation program is called may refer to the various embodiments of the automobile acceleration noise evaluation method of the present application, and will not be described in detail here.
[0099] Those skilled in the art will understand that Figure 7 The hardware structure shown in the figure does not constitute a limitation on the present application, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0100] In a third aspect, an embodiment of the present application also provides a computer-readable storage medium.
[0101] The computer-readable storage medium of the present application stores a vehicle acceleration noise evaluation program, wherein when the vehicle acceleration noise evaluation program is executed by a processor, the steps of the vehicle acceleration noise evaluation method as described above are implemented.
[0102] Among them, the method implemented when the automobile acceleration noise evaluation program is executed can refer to the various embodiments of the automobile acceleration noise evaluation method of the present application, and will not be repeated here.
[0103] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.
[0104] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.
[0105] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.
[0106] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.
[0107] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; the “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.
[0108] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.
[0109] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.
[0110] The above are only specific implementations of the embodiments of the present invention, but the protection scope of the embodiments of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the embodiments of the present invention, and these modifications or replacements should be included in the protection scope of the embodiments of the present invention. Therefore, the protection scope of the embodiments of the present invention shall be based on the protection scope of the claims.
Claims
1. A method for evaluating automobile acceleration noise, characterized in that: The method comprises the following steps: Acquire a noise signal set within an engine speed range, and segment the noise signal set to form a plurality of noise signal data segments; Determine the MFCC features of each noise signal data segment, and determine the MFCC evaluation result according to each MFCC feature and its corresponding engine speed; The timbre mutation evaluation result is determined based on the similarity of the MFCC features of adjacent segments; The evaluation results of automobile acceleration noise are determined based on the MFCC evaluation results and the timbre mutation evaluation results.
2. The vehicle acceleration noise evaluation method according to claim 1, characterized in that: The process of obtaining a noise signal set within an engine speed range includes: respectively acquiring an engine lower speed limit closest to a lower speed limit of the engine speed range and an engine upper speed limit closest to an upper speed limit of the engine speed range; forming an engine speed set according to an upper engine speed limit, a lower engine speed limit, and other engine speeds between the upper engine speed limit and the lower engine speed limit; The noise signals corresponding to all engine speeds in the engine speed set are taken as the noise signal set.
3. The vehicle acceleration noise evaluation method according to claim 1, characterized in that: The calculation formula of the segmentation time length T for segmenting the noise signal set is: T=N*Δt, N represents the number of sampling points of the noise signal in each noise signal set after segmentation, Δt represents the noise sampling interval, Δt=1 / fs, fs is the noise sampling frequency.
4. The vehicle acceleration noise evaluation method according to claim 3, characterized in that: The process of segmenting the noise signal set to form a plurality of noise signal data segments includes: Determine the total duration according to the minimum sampling time and the maximum sampling time in the noise signal set, and determine the time range of each segment according to the total duration and the segment duration; The noise signal at the sampling moment belonging to each time range is taken as the noise signal data of that section.
5. The vehicle acceleration noise evaluation method according to claim 1, characterized in that: The process of determining the MFCC features of each noise signal data segment and determining the MFCC evaluation result according to each MFCC feature and its corresponding engine speed includes: Determine the MFCC features of each segment of noise signal data, and calculate the MFCC amplitude of each MFCC feature vector; Based on each MFCC amplitude and its corresponding engine speed, an evaluation curve of MFCC amplitude changing with engine speed is formed; The MFCC evaluation result is determined according to the evaluation curve.
6. The vehicle acceleration noise evaluation method according to claim 5, characterized in that: The process of forming an evaluation curve of MFCC amplitude changing with engine speed according to each MFCC amplitude and its corresponding engine speed includes: The engine speed and MFCC amplitude are used as the horizontal and vertical coordinates respectively; Determine the coordinate point according to each MFCC amplitude and its corresponding engine speed; Connect all coordinate points to obtain the evaluation curve of MFCC amplitude changing with engine speed; The process of determining the MFCC evaluation result according to the evaluation curve includes: generating a target line corresponding to the vehicle acceleration noise target requirement, judging whether the evaluation curve intersects with the target line, and if so, determining that the MFCC evaluation result is not passed, otherwise determining that the MFCC evaluation result is passed.
7. The vehicle acceleration noise evaluation method according to claim 1, characterized in that: The process of determining the timbre mutation evaluation result according to the similarity of adjacent MFCC features includes: calculating the cosine similarity of the MFCC feature vectors of adjacent segmented noise signal data; judging whether the number of times the cosine similarity is less than a reference threshold exceeds a specified value, and if so, determining that the timbre mutation evaluation result fails, otherwise determining that the timbre mutation evaluation result passes.
8. The vehicle acceleration noise evaluation method according to any one of claims 1 to 7, characterized in that: The process of determining the automobile acceleration noise evaluation result according to the MFCC evaluation result and the timbre mutation evaluation result includes: judging whether both the MFCC evaluation result and the timbre mutation evaluation result are passed; if so, determining that the automobile acceleration noise evaluation result is passed; otherwise, determining that the automobile acceleration noise evaluation result is failed.
9. A vehicle acceleration noise evaluation device, characterized in that: The automobile acceleration noise evaluation device comprises a processor, a memory, and an automobile acceleration noise evaluation program stored in the memory and executable by the processor, wherein when the automobile acceleration noise evaluation program is executed by the processor, the steps of the automobile acceleration noise evaluation method according to any one of claims 1 to 8 are implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a vehicle acceleration noise evaluation program, wherein when the vehicle acceleration noise evaluation program is executed, the steps of the vehicle acceleration noise evaluation method according to any one of claims 1 to 8 are implemented.
Citation Information
Patent Citations
Speech evaluation method and device
CN110600052A
Vehicle door closing sound quality evaluation method and system
CN114220455A
Automobile acceleration sound quality evaluation method, device and equipment and storage medium
CN115307721A
