Sound quality evaluation method, computer equipment and storage medium

Abstract

The present invention relates to the field of sound processing, and discloses a sound quality evaluation method, a computer device, and a storage medium, wherein the method comprises: obtaining sound data and subjective evaluation words; obtaining sound parameters matching the subjective evaluation words; determining characteristic values ​​of the sound parameters according to the sound data; and determining evaluation data of the sound data on the subjective evaluation words according to the characteristic values. The present invention can improve the reproducibility and accuracy of evaluation data; at the same time, it can also reduce the evaluation time for obtaining evaluation data, improve evaluation efficiency, and reduce evaluation costs.

Description

Technical Field

[0001] The present invention relates to the field of sound processing, and in particular to a sound quality evaluation method, a computer device and a storage medium. Background Art

[0002] With the popularization of automobiles, providing users with a highly comfortable automobile product is an important means for automobile manufacturers to improve their market competitiveness. Among them, the sound quality of the vehicle is an important indicator of comfort.

[0003] In order to improve the sound quality of vehicles, it is necessary to collect evaluation feedback data on vehicle sound effects. Existing evaluation feedback data mainly adopts subjective evaluation method, but this method consumes a lot of manpower and time, and the evaluation data will also be affected by the evaluation status of the evaluator. Therefore, it is necessary to find a new sound quality evaluation method to reduce the evaluation time and cost of obtaining evaluation data. Summary of the invention

[0004] Based on this, it is necessary to provide a sound quality evaluation method, device, computer equipment and storage medium to address the above technical problems, so as to reduce the evaluation time for obtaining evaluation data and reduce the evaluation cost.

[0005] A sound quality evaluation method, comprising:

[0006] Acquire sound data and subjective evaluation words;

[0007] Acquiring sound parameters matching the subjective evaluation words;

[0008] Determining a characteristic value of the sound parameter according to the sound data;

[0009] Evaluation data of the sound data on the subjective evaluation words is determined based on the feature value.

[0010] A computer device comprises a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned sound quality evaluation method when executing the computer-readable instructions.

[0011] A computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions implement the above-mentioned sound quality evaluation method when executed by a processor.

[0012] The above-mentioned sound quality evaluation method, device, computer equipment and storage medium obtain the evaluation object (sound data) and evaluation attributes (subjective evaluation words) by acquiring sound data and subjective evaluation words. The sound parameters matching the subjective evaluation words are obtained. Here, the sound parameters are objective parameters, and the matching relationship between the sound parameters and the subjective evaluation words is determined based on a preset regression model. The characteristic values ​​of the sound parameters are determined according to the sound data. Here, the characteristic values ​​of the sound parameters are extracted from the sound data, and the obtained characteristic values ​​are unique. The evaluation data of the sound data on the subjective evaluation words is determined according to the characteristic values ​​to obtain the final required evaluation data. Here, the evaluation data can replace the evaluation data obtained by the subjective evaluation method to a certain extent, reduce the evaluation time of the evaluation data, and improve the reproducibility and accuracy of the evaluation data. The present invention can reduce the evaluation time for obtaining evaluation data and reduce the evaluation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0014] Figure 1 is a schematic diagram of an application environment of a sound quality evaluation method in one embodiment of the present invention;

[0015] Figure 2 A comparison between the evaluation score obtained by the subjective evaluation method and the overall sound quality evaluation data obtained by the sound quality evaluation method provided by this embodiment;

[0016] Figure 3 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] In one embodiment, if Figure 1 As shown, a sound quality evaluation method is provided, comprising the following steps S10-S40.

[0019] S10. Acquire sound data and subjective evaluation words.

[0020] Here, the sound data can be audio recorded in a car. Schematically, in a car in a semi-anechoic chamber, a simulated human head model is placed in the front driver's seat or the right side of the rear seat, and the seat height and position are adjusted. The test sound sample is played, and the sound data is collected by the acquisition device set in the ears of the simulated human head model. Here, the height of the simulated human head model is 70 cm, the distance between the head and the seat is 4 cm, and the seat angle is 115°.

[0021] Subjective evaluation words can refer to people's subjective feelings used to evaluate a certain aspect of sound. Indicatively, subjective evaluation words can be strength, brightness, vocals (similar to the pleasantness of human voice), or the overall evaluation of the sound. Here, the subjective evaluation words are the subjective characteristics of the sound that need to be evaluated.

[0022] S20: Acquire sound parameters that match the subjective evaluation words.

[0023] In this embodiment, it is necessary to perform a sound parameter (objective) analysis and a subjective sound quality evaluation (evaluated by subjective evaluation words) on the in-vehicle audio system in advance, and then use the regression analysis method to determine the mapping relationship (that is, the matching relationship) between the two. Furthermore, the sound parameters that match the subjective evaluation words can be obtained based on the mapping relationship. Generally, the number of sound parameters that match the subjective evaluation words is two or more. Here, the regression analysis method uses multiple linear regression to establish a model, and uses the backward method to screen the sound parameters of the subjective evaluation words. Schematically, the sound parameters that match the subjective evaluation word "intensity" include low-frequency extension parameters and low-frequency harmonic distortion parameters.

[0024] S30. Determine a characteristic value of the sound parameter according to the sound data.

[0025] Here, the sound parameter is an objective parameter, and the sound data can be processed based on the definition of the sound parameter to obtain the corresponding characteristic value of the sound parameter. In some cases, the sound parameter reflects the sound characteristics of the sound data in a specific frequency band.

[0026] S40: Determine evaluation data of the sound data based on the subjective evaluation words according to the feature value.

[0027] In this embodiment, after obtaining the characteristic value of the sound parameter, since there is a mapping relationship between the sound parameter and the subjective evaluation word, the evaluation data of the sound data on the subjective evaluation word can be calculated based on the mapping relationship. Schematically, a subjective evaluation word is intensity (A), and its matching sound parameters are low-frequency extension parameter (LFX) and low-frequency harmonic distortion parameter (THD L ). When determining LFX and THD L After the value of, then A = f (LFX, THDL ), where f is used to represent A and (LFX, THD L ) between them.

[0028] In steps S10-S40, the evaluation object (sound data) and the evaluation attribute (subjective evaluation word) are obtained by acquiring sound data and subjective evaluation words. The sound parameters matching the subjective evaluation words are obtained. Here, the sound parameters are objective parameters, and the matching relationship between the sound parameters and the subjective evaluation words is determined based on a preset regression model. The characteristic values ​​of the sound parameters are determined according to the sound data. Here, the characteristic values ​​of the sound parameters are extracted from the sound data, and the obtained characteristic values ​​are unique. The evaluation data of the sound data on the subjective evaluation words is determined according to the characteristic values ​​to obtain the final required evaluation data. Here, the evaluation data can replace the evaluation data obtained by the subjective evaluation method to a certain extent, reduce the evaluation time of the evaluation data, and improve the reproducibility and accuracy of the evaluation data.

[0029] Optionally, if the subjective evaluation word is strength, the sound parameters matching the subjective evaluation word include a low-frequency extension parameter and a low-frequency harmonic distortion parameter;

[0030] Step S40, i.e. determining the evaluation data of the sound data on the subjective evaluation words according to the feature values, comprises:

[0031] The characteristic value is processed by a strength formula to generate strength evaluation data, and the strength formula includes:

[0032] A=a+b*LFX+c*THD L ;

[0033] Among them, A is the dynamic evaluation data, LFX is the characteristic value of the low-frequency extension parameter, THD L is the characteristic value of the low-frequency harmonic distortion parameter, and a, b, c are constant coefficients.

[0034] In this embodiment, the historical sound parameters and subjective evaluation data are analyzed based on the regression model, and the low-frequency extension parameter and the low-frequency harmonic distortion parameter are determined from multiple sound parameters to have a mapping relationship with the intensity. In one example, by solving the regression model, the specific values ​​of the constant coefficients in the intensity formula can be obtained, specifically: a=1.678, b=-0.119, c=-0.009.

[0035] Optionally, step S30, i.e., determining the characteristic value of the sound parameter according to the sound data, includes:

[0036] The characteristic value of the low-frequency extension parameter is calculated by a low-frequency extension formula, and the low-frequency extension formula is:

[0037]

[0038] Among them, y REF1 is a first reference amplitude, the first reference amplitude being the sum of an average amplitude of the audio in a first reference frequency band and a preset specified value; the frequency range of the first reference frequency band is 300 Hz to 10 kHz; Indicates that in the frequency band below 300 Hz, the first amplitude is lower than y REF1 The frequency of the value 6dB;

[0039] The characteristic value of the low-frequency harmonic distortion parameter is calculated by a low-frequency harmonic distortion formula, and the low-frequency harmonic distortion formula is:

[0040]

[0041] Among them, N 50-300Hz is the total number of frequency points measured in the frequency range of 50Hz to 300Hz, f n1 is the frequency value corresponding to the n1th frequency point, n1 is an integer, and 0<n1≤N 50-300Hz ;THD(f n1 ) is the same as f n1 The corresponding eigenvalue of the harmonic distortion parameter.

[0042] In this embodiment, refers to the amplitude lower than y REF1 The value is 6dB, which is less than and closest to the frequency of 300Hz. In other words, In the frequency band below 300Hz, all amplitudes below y REF1 The frequency with the highest frequency value among the frequencies with a value of 6dB. When calculating the eigenvalue of the low-frequency extension parameter, by taking the logarithm, it is convenient to construct a linear mapping relationship between subjective data and objective data in the regression model. Due to the enhancement of the low frequency by the in-car audio equalizer, a preset specified value is added on the basis of the average amplitude in the first reference frequency band. The preset specified value is an empirical value, generally +10dB. The first reference frequency band can generally be selected from 300Hz to 10kHz. When used to evaluate the in-car sound field, the first reference frequency band can be set to 300Hz to 3kHz.

[0043] Here, the characteristic value of the low-frequency harmonic distortion parameter can be the average value of all total harmonic distortions (THD) in the low frequency band (50-300Hz). The value of n can be set as needed. Indicatively, sampling can be performed every 10Hz. Total harmonic distortion (THD) refers to the additional harmonic components of the output signal caused by nonlinear elements when the audio signal source passes through the power amplifier compared to the input signal.

[0044] Optionally, if the subjective evaluation word is brightness, the sound parameters matching the subjective evaluation word include a high-frequency extension parameter and a high-frequency quality parameter;

[0045] Step S40, i.e. determining the evaluation data of the sound data on the subjective evaluation words according to the feature values, comprises:

[0046] The characteristic value is processed by a brightness formula to generate brightness evaluation data, and the brightness formula includes:

[0047] B = d + e * HFQ + f * HFX;

[0048] Among them, B is the brightness evaluation data, HFX is the characteristic value of the high-frequency extension parameter, HFQ is the characteristic value of the high-frequency quality parameter, and d, e, and f are constant coefficients.

[0049] In this embodiment, the matching relationship between the subjective evaluation word brightness and the two sound parameters of the high-frequency extension parameter and the high-frequency quality parameter is determined based on the analysis result of the regression model. In one example, by solving the regression model, the specific values ​​of the constant coefficients in the brightness formula can be obtained, specifically: d = 6.378, e = -0.359, f = 0.081.

[0050] Optionally, determining the characteristic value of the sound parameter according to the sound data includes:

[0051] The characteristic value of the high-frequency extension parameter is calculated by a high-frequency extension formula, and the high-frequency extension formula is:

[0052]

[0053] Among them, y REF2 is a second reference amplitude, the second reference amplitude is an average amplitude of audio in a second reference frequency band, the second reference frequency band is 300 to 3 kHz, Indicates that in the frequency band above 5kHz, the first REF2 6dB frequency;

[0054] The characteristic value of the high-frequency quality parameter is calculated by a high-frequency quality formula, and the high-frequency quality formula is:

[0055]

[0056] Where N1 is the total number of 1 / 20 octave bands between 5kHz and the highest frequency defined by HFX, It represents the average amplitude corresponding to the n2th octave band, where n2 is an integer and 0<n2≤N1.

[0057] In this embodiment, refers to the amplitude lower than y REF2 The value is 6dB, which is greater than and closest to the frequency of 5kHz. In other words, In the frequency band above 5kHz, all amplitudes below y REF2 The frequency with the lowest frequency value among the frequencies with a value of 6dB. When calculating the characteristic value of the high frequency extension parameter (HFX), taking the logarithm facilitates the construction of a linear mapping relationship between subjective data and objective data in the regression model. HFX is theoretically positively correlated with the subjective evaluation score.

[0058] The highest frequency defined by HFX can be the highest sampling frequency of the sound data. For example, if the highest sampling frequency of the sound data is 30kHz, the highest frequency defined by HFX is 30kHz. The high frequency quality parameter (HFQ) is used to quantify the amplitude response deviation in the high frequency area between 5kHz and the high frequency cutoff frequency. Here, the HFX and HFQ parameters are related to the subjective brightness.

[0059] Optionally, if the subjective evaluation word is a human voice, the sound parameters matching the subjective evaluation word include narrowband deviation, spectrum flatness and intermediate frequency harmonic distortion parameters;

[0060] Step S40, i.e. determining the evaluation data of the sound data on the subjective evaluation words according to the feature values, comprises:

[0061] The characteristic value is processed by a vocal formula to generate vocal evaluation data, wherein the vocal formula includes:

[0062] C=g+h*NBD+i*SPF+j*THD M ;

[0063] Among them, C is the human voice evaluation data, NBD is the characteristic value of narrowband deviation, SPF is the characteristic value of spectrum flatness, THD is M is the characteristic value of the intermediate frequency harmonic distortion parameter, and g, h, i, and j are constant coefficients.

[0064] In this embodiment, the historical sound parameters and subjective evaluation data are analyzed based on the regression model, and the three sound parameters of narrowband deviation, spectrum flatness and intermediate frequency harmonic distortion parameters are determined from multiple sound parameters to have a mapping relationship with human voice. In one example, by solving the regression model, the specific values ​​of the constant coefficients in the dynamics formula can be obtained, which are: g = 14.822, h = -2.113, i = 1.523, g = -3.824.

[0065] Optionally, step S30, i.e., determining the characteristic value of the sound parameter according to the sound data, includes:

[0066] The characteristic value of the narrowband deviation is calculated by a narrowband deviation formula, and the narrowband deviation formula includes:

[0067]

[0068] Where N2 is the total number of 1 / 2 octave bands between 100 Hz and 12 kHz, is the average amplitude value in the n3th 1 / 2 octave band, y b It is the frequency band b in the n3th 1 / 2 octave band f The amplitude value of , n3 is an integer, and 0<n≤N2;

[0069] The characteristic value of the spectrum flatness is calculated by the spectrum flatness formula, and the spectrum flatness formula is:

[0070]

[0071] Among them, s k is the value of the signal amplitude spectrum at frequency point k, K is the upper limit frequency of a preset frequency band, and the preset frequency band range includes 300 Hz to 20 kHz;

[0072] The characteristic value of the intermediate frequency harmonic distortion parameter is calculated by the intermediate frequency harmonic distortion formula, and the intermediate frequency harmonic distortion formula is:

[0073]

[0074] Among them, N 300-3000Hz is the total number of frequency points measured in the 300Hz to 3000Hz frequency band, f n4 is the frequency value corresponding to the n4th frequency point, THD(f n4 ) is the same as f n4 The corresponding eigenvalue of the harmonic distortion parameter.

[0075] In this embodiment, when calculating the characteristic value of narrowband deviation, the average absolute deviation in each 1 / 2 octave band is based on 10 equally logarithmically spaced amplitude data points sampled in the octave band. The higher the NBD value, the greater the amplitude deviation in the narrowband. Here, the characteristic value of NBD is negatively correlated with the subjective evaluation score.

[0076] Spectral flatness (SPF) is the ratio of the geometric mean to the arithmetic mean of the signal power spectrum. This ratio is between 0 and 1. For a white noise, the calculated result is 1; for a pure tone signal, the calculated result is 0. The closer the SPF is to 1, the higher the flatness. Taking the logarithm to expand the value range, the larger the result after taking the logarithm (the closer to 0), the higher the flatness. The preset frequency band range of SPF is 300Hz-20kHz. SPF is positively correlated with the subjective evaluation score, and its characteristic value is related to the human voice being evaluated (that is, the SPF is different for different people's pronunciation).

[0077] The characteristic value of the intermediate frequency harmonic distortion parameter may be the average value of all total harmonic distortions (THD) in the intermediate frequency band (300 Hz to 3000 Hz). The value of n4 may be set as required.

[0078] Optionally, if the subjective evaluation word is the overall sound quality, the sound parameters matching the subjective evaluation word include low-frequency extension parameter, low-frequency harmonic distortion parameter, high-frequency extension parameter, high-frequency quality parameter, narrowband deviation, spectrum flatness and intermediate-frequency harmonic distortion parameter;

[0079] The step of determining the evaluation data of the sound data on the subjective evaluation words according to the feature value comprises:

[0080] The characteristic value is processed by an overall sound quality formula to generate overall sound quality evaluation data, wherein the overall sound quality formula includes:

[0081] S=k+l*LFX+m*THD L +n*HFQ+o*HFX+p*NBD+q*SPF+r*THD M ;

[0082] Among them, S is the overall sound quality evaluation data,

[0083] LFX is the characteristic value of the low-frequency extension parameter,

[0084] THD L is the characteristic value of the low-frequency harmonic distortion parameter,

[0085] HFX is the characteristic value of the high frequency extension parameter,

[0086] HFQ is the characteristic value of the high frequency quality parameter,

[0087] NBD is the characteristic value of narrowband deviation,

[0088] SPF is the characteristic value of spectrum flatness,

[0089] THD M is the characteristic value of the intermediate frequency harmonic distortion parameter,

[0090] k, l, m, n, o, p, q, and r are constant coefficients.

[0091] In this embodiment, the historical sound parameters and subjective evaluation data are analyzed based on the regression model, and the sound parameters with mapping relationships with strength, brightness, and human voice are determined respectively. The overall sound quality evaluation data can be the weighted mean of the evaluation data of these three subjective evaluation words. In one example, the weight coefficients of the evaluation data of the three subjective evaluation words are equal. At this time, the constant coefficients in the overall sound quality formula are: k = 7.626 (i.e. k = 1 / 3 (a + d + g), and so on), l = -0.04 (i.e. l ​​= 1 / 3b), m = -0.003, n = -0.12, o = 0.027, p = -0.704, q = 0.508, r = -1.275.

[0092] The calculation process of the characteristic values ​​of each sound parameter can refer to the calculation process of the previous embodiment, which will not be repeated here.

[0093] In order to determine the relative importance of sound parameters in the regression model, the regression coefficients were standardized and sorted by absolute value. The results are shown in Table 1.

[0094] Table 1 Standardized coefficients of each sound parameter in the regression model

[0095] Predictor variables Standardized coefficient SPF 0.245 HFX 0.212 NBD -0.165 <![CDATA[THD M ]]> -0.121 HFQ -0.106 LFX -0.006

[0096] In Table 1, SPF and HFX are most important to the dependent variable. A positive sign indicates a positive correlation with the dependent variable, and a negative sign indicates a negative correlation with the dependent variable. Therefore, we can improve SPF and HFX and reduce THD. M The subjective evaluation score can be improved by Figure 2 As shown, Figure 2 It is a comparison between the evaluation score obtained by the subjective evaluation method and the overall sound quality evaluation data obtained by adopting the sound quality evaluation method provided by this embodiment.

[0097] The regression model was validated by using the leave-one-out cross validation method. One sample was selected from the six samples as the test sample, and the remaining five samples were used as training samples to obtain the regression model. The prediction effect of the regression model on the test sample was analyzed. The test samples were repeatedly selected until each sample was used as a test sample. The validation data is shown in Table 2.

[0098] Table 2 Validation data of regression model

[0099]

[0100] In Table 2, the prediction relative error is basically within 15%. The average absolute relative error is 9.93%, which is less than 10%. The root mean square error RMSE is 0.81, which is less than 0.9, indicating that the regression model has a certain prediction effect.

[0101] In one example, two cars of the same model and the same audio hardware level are subjected to staged sound tests (including before and after tuning) to obtain corresponding sound data. The sound data are evaluated using the subjective evaluation method and the evaluation method provided by the embodiment of the present invention, as shown in Table 3.

[0102] Table 3 Evaluation data obtained by different evaluation methods

[0103]

[0104]

[0105] S1 is the evaluation score obtained by the subjective evaluation method, and S2 is the evaluation score obtained by the evaluation method provided in the embodiment of the present invention.

[0106] In Table 3, after tuning, the evaluation scores obtained by the evaluation method provided by the embodiment of the present invention all increased, which is consistent with the evaluation scores obtained by the subjective evaluation method. Among them, the evaluation score of the embodiment of the present invention in terms of strength has increased significantly, and correspondingly, the evaluation score of the subjective evaluation method for strength has also increased significantly. This shows that the sound quality evaluation method provided by the embodiment of the present invention has high evaluation data accuracy in terms of strength. In terms of brightness and vocals, although the increase in the evaluation scores obtained by the embodiment of the present invention is not as good as the evaluation scores of the subjective evaluation method, there will be no negative change. Therefore, in general, the evaluation method provided by the embodiment of the present invention can be used to evaluate the sound quality before and after vehicle tuning.

[0107] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0108] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer-readable instructions. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer-readable instructions are executed by the processor, a sound quality evaluation method is implemented.

[0109] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor executes the computer-readable instructions to implement the following steps:

[0110] Acquire sound data and subjective evaluation words;

[0111] Acquiring sound parameters matching the subjective evaluation words;

[0112] Determining a characteristic value of the sound parameter according to the sound data;

[0113] Evaluation data of the sound data on the subjective evaluation words is determined based on the feature value.

[0114] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The readable storage media store computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the following steps are implemented:

[0115] Acquire sound data and subjective evaluation words;

[0116] Acquiring sound parameters matching the subjective evaluation words;

[0117] Determining a characteristic value of the sound parameter according to the sound data;

[0118] Evaluation data of the sound data on the subjective evaluation words is determined based on the feature value.

[0119] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile computer-readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0120] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0121] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A sound quality evaluation method, characterized in that: include: Acquire sound data and subjective evaluation words; Acquiring sound parameters matching the subjective evaluation words; Determining a characteristic value of the sound parameter according to the sound data; determining evaluation data of the sound data on the subjective evaluation words according to the feature value; If the subjective evaluation word is strength, the sound parameters matching the subjective evaluation word include a low-frequency extension parameter and a low-frequency harmonic distortion parameter; Determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic value includes: determining the strength evaluation data of the sound data based on a regression model according to the characteristic value of the low-frequency extension parameter and the characteristic value of the low-frequency harmonic distortion parameter.

2. The sound quality evaluation method according to claim 1, wherein: The step of determining the evaluation data of the sound data on the subjective evaluation words according to the feature value comprises: The characteristic value is processed by a strength formula to generate strength evaluation data, and the strength formula includes: Among them, A is the strength evaluation data, LFX is the characteristic value of the low-frequency extension parameter, is the characteristic value of the low-frequency harmonic distortion parameter, and a, b, c are constant coefficients.

3. The sound quality evaluation method according to claim 2, wherein: The determining the characteristic value of the sound parameter according to the sound data comprises: The characteristic value of the low-frequency extension parameter is calculated by a low-frequency extension formula, and the low-frequency extension formula is: in, is a first reference amplitude, wherein the first reference amplitude is the sum of an average amplitude of the audio in a first reference frequency band and a preset specified value; the frequency range of the first reference frequency band is 300 Hz - 10 kHz; Indicates that in the frequency band below 300Hz, the first amplitude is lower than The frequency of the value 6dB; The characteristic value of the low-frequency harmonic distortion parameter is calculated by a low-frequency harmonic distortion formula, and the low-frequency harmonic distortion formula is: in, is the total number of frequency points measured in the frequency range of 50 Hz to 300 Hz; is the frequency value corresponding to the n1th frequency point, n1 is an integer, and 0 <n1≤ ; For The corresponding eigenvalue of the harmonic distortion parameter.

4. The sound quality evaluation method according to claim 1, wherein: If the subjective evaluation word is brightness, the sound parameters matching the subjective evaluation word include a high-frequency extension parameter and a high-frequency quality parameter; the highest frequency defined by the high-frequency extension parameter is the highest sampling frequency of the sound data; the high-frequency quality parameter is used to quantify the amplitude response deviation of the high-frequency area between 5 kHz and the high-frequency cutoff frequency; The step of determining the evaluation data of the sound data on the subjective evaluation words according to the feature value comprises: The characteristic value is processed by a brightness formula to generate brightness evaluation data, and the brightness formula includes: Among them, B is the brightness evaluation data; is the characteristic value of the high-frequency extension parameter; is the characteristic value of the high-frequency quality parameter; d, e, f are constant coefficients.

5. The sound quality evaluation method according to claim 4, characterized in that: The determining the characteristic value of the sound parameter according to the sound data comprises: The characteristic value of the high-frequency extension parameter is calculated by a high-frequency extension formula, and the high-frequency extension formula is: in, is a second reference amplitude, the second reference amplitude is an average amplitude of audio in a second reference frequency band, and the second reference frequency band is 300 ~ 3kHz; Indicates that in the frequency band above 5 kHz, the first amplitude is lower than The frequency of the value 6dB; The characteristic value of the high-frequency quality parameter is calculated by a high-frequency quality formula, and the high-frequency quality formula is: where N1 is the total number of 1 / 20 octave bands between 5 kHz and the highest frequency defined by HFX; represents the average amplitude corresponding to the n2th octave band, where n2 is an integer and 0 <n2≤N1。 6. The sound quality evaluation method according to claim 1, wherein: If the subjective evaluation word is a human voice, the sound parameters matching the subjective evaluation word include narrowband deviation, spectrum flatness and intermediate frequency harmonic distortion parameters; The step of determining the evaluation data of the sound data on the subjective evaluation words according to the feature value comprises: The characteristic value is processed by a vocal formula to generate vocal evaluation data, wherein the vocal formula includes: Among them, C is the voice evaluation data, is the eigenvalue of the narrowband deviation, is the characteristic value of spectrum flatness, is the characteristic value of the intermediate frequency harmonic distortion parameter, and g, h, i, and j are constant coefficients.

7. The sound quality evaluation method according to claim 6, wherein: The determining the characteristic value of the sound parameter according to the sound data comprises: The characteristic value of the narrowband deviation is calculated by a narrowband deviation formula, and the narrowband deviation formula includes: in, N 2 is the total number of 1 / 2 octave bands between 100 Hz and 12 kHz, It is n The average amplitude value in 3 1 / 2 octave bands, It is n Frequency bands in 3 1 / 2 octave bands b f The amplitude value, n3 is an integer, and 0 <n3≤N2; The characteristic value of the spectrum flatness is calculated by the spectrum flatness formula, and the spectrum flatness formula is: in, is the signal amplitude spectrum at the frequency point The value on The upper limit frequency of the preset frequency band range includes 300 Hz ~ 20 kHz; The characteristic value of the intermediate frequency harmonic distortion parameter is calculated by the intermediate frequency harmonic distortion formula, and the intermediate frequency harmonic distortion formula is: in, is the total number of frequency points measured in the 300Hz~3000Hz frequency band, is the frequency value corresponding to the n4th frequency point, For The corresponding eigenvalue of the harmonic distortion parameter.

8. The sound quality evaluation method according to claim 1, wherein: If the subjective evaluation word is the overall sound quality, the sound parameters matching the subjective evaluation word include low-frequency extension parameters, low-frequency harmonic distortion parameters, high-frequency extension parameters, high-frequency quality parameters, narrowband deviation, spectrum flatness and intermediate-frequency harmonic distortion parameters; the highest frequency defined by the high-frequency extension parameter is the highest sampling frequency of the sound data; the high-frequency quality parameter is used to quantify the amplitude response deviation of the high-frequency area between 5 kHz and the high-frequency cutoff frequency; The step of determining the evaluation data of the sound data on the subjective evaluation words according to the feature value comprises: The characteristic value is processed by an overall sound quality formula to generate overall sound quality evaluation data, wherein the overall sound quality formula includes: Among them, S is the overall sound quality evaluation data, LFX is the characteristic value of the low-frequency extension parameter, is the characteristic value of the low-frequency harmonic distortion parameter, is the eigenvalue of the high-frequency extension parameter, is the eigenvalue of the high-frequency quality parameter, is the eigenvalue of the narrowband deviation, is the characteristic value of spectrum flatness, is the characteristic value of the intermediate frequency harmonic distortion parameter, k, l, m, n, o, p, q, and r are constant coefficients.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the sound quality evaluation method according to any one of claims 1 to 8 is implemented.

10. One or more readable storage media storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the sound quality evaluation method according to any one of claims 1 to 8.

Images