CN114067841A - Sound quality evaluation method, computer device, and storage medium - Google Patents
Sound quality evaluation method, computer device, and storage medium Download PDFInfo
- Publication number
- CN114067841A CN114067841A CN202010745297.XA CN202010745297A CN114067841A CN 114067841 A CN114067841 A CN 114067841A CN 202010745297 A CN202010745297 A CN 202010745297A CN 114067841 A CN114067841 A CN 114067841A
- Authority
- CN
- China
- Prior art keywords
- frequency
- sound
- evaluation
- parameter
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 29
- 238000011156 evaluation Methods 0.000 claims abstract description 167
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000003595 spectral effect Effects 0.000 claims description 14
- 238000001228 spectrum Methods 0.000 claims description 5
- 238000001303 quality assessment method Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 description 10
- 230000000875 corresponding effect Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/60—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for measuring the quality of voice signals
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to the field of sound processing, and discloses a sound quality evaluation method, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring sound data and subjective evaluation words; acquiring sound parameters matched with the subjective evaluation words; determining a characteristic value of the sound parameter according to the sound data; and determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values. The invention can improve the reproducibility and accuracy of the evaluation data; meanwhile, the evaluation time for obtaining the evaluation data can be reduced, the evaluation efficiency is improved, and the evaluation cost is reduced.
Description
Technical Field
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
With the popularization of automobiles, an automobile product with high comfort level is provided for users, and the automobile product is an important means for automobile manufacturers to improve market competitiveness. The sound effect quality of the vehicle is an important index for representing the comfort level.
In order to improve the sound effect quality of the vehicle, evaluation feedback data on the sound effect of the vehicle needs to be collected. The existing evaluation feedback data mainly adopts a subjective evaluation method, but the method consumes a large amount of manpower and time, and meanwhile, the evaluation data is also influenced by the evaluation state of an evaluator. Therefore, a new sound quality evaluation method is needed to be found, so that the evaluation time for acquiring evaluation data is reduced, and the evaluation cost is reduced.
Disclosure of Invention
In view of the above, it is desirable to provide a sound quality evaluation method, device, computer device and storage medium for reducing evaluation time for acquiring evaluation data and reducing evaluation cost.
An acoustic quality evaluation method comprising:
acquiring sound data and subjective evaluation words;
acquiring sound parameters matched with the subjective evaluation words;
determining a characteristic value of the sound parameter according to the sound data;
and determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values.
A computer device comprising a memory, a processor and computer readable instructions stored in the memory and executable on the processor, the processor implementing the sound quality assessment method when executing the computer readable instructions.
A computer-readable storage medium storing computer-readable instructions which, when executed by a processor, implement the sound quality evaluation method described above.
The sound quality evaluation method, the sound quality evaluation device, the computer equipment and the storage medium acquire the sound data and the subjective evaluation words to obtain the evaluation objects (sound data) and the evaluation attributes (subjective evaluation words). And acquiring sound parameters matched with the subjective evaluation words, wherein the sound parameters are objective parameters, and the matching relation between the sound parameters and the subjective evaluation words is determined based on a preset regression model. And determining the characteristic value of the sound parameter according to the sound data, wherein the characteristic value of the sound parameter is extracted from the sound data, and the obtained characteristic value has uniqueness. And determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values to obtain the finally required evaluation data, wherein the evaluation data can replace the evaluation data obtained by a subjective evaluation method to a certain extent, so that the evaluation time of the evaluation data is reduced, and the reproducibility and the accuracy of the evaluation data are improved. The invention can reduce the evaluation time for obtaining the evaluation data and reduce the evaluation cost.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic diagram of an application environment of a sound quality evaluation method according to an embodiment of the present invention;
fig. 2 is 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 in this embodiment;
FIG. 3 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In one embodiment, as shown in FIG. 1, a method for evaluating sound quality is provided, which includes the following steps S10-S40.
And S10, acquiring sound data and subjective evaluation words.
Here, the sound data may be audio recorded in a car. Schematically, in an automobile in a semi-anechoic chamber, a simulated human head model is placed at the position on the right side of a front-row main driver or a rear row, and the height and the position of a seat are adjusted. And playing a test sound sample, and collecting sound data by collecting equipment arranged in the two ears of the simulated human head model. Here, the dummy head model was 70 cm in height, the head was 4 cm away from the seat, and the seat angle was 115 °.
Subjective terms may refer to a subjective feeling that a person uses to rate some aspect of sound. Illustratively, the subjective evaluation words may be dynamics, brightness, human voice (similar to the pleasure degree of human voice), or may be the overall evaluation of sound. Here, the subjective evaluation word is a subjective feature of sound to be evaluated.
And S20, acquiring the sound parameters matched with the subjective evaluation words.
In this embodiment, it is necessary to perform objective analysis and subjective sound quality evaluation (evaluation by subjective evaluation terms) on the in-vehicle sound system in advance, and then determine the mapping relationship (i.e., the matching relationship) between the two by using a regression analysis method. Further, the sound parameter matching the subjective evaluation word may be acquired based on the mapping relationship. In general, the number of sound parameters matched with the subjective evaluation word is two or more. Here, the regression analysis method uses a multiple linear regression modeling and a backward method to screen the sound parameters of the subjective evaluation words. Illustratively, the sound parameters matched with the subjective evaluation word "strength" include a low-frequency extension parameter and a low-frequency harmonic distortion parameter.
And S30, determining the characteristic value of the sound parameter according to the sound data.
Here, the sound parameter belongs to an objective parameter, and the sound data may be processed based on the definition of the sound parameter to obtain a feature value of the corresponding sound parameter. In some cases, the sound parameters reflect the sound characteristics of the sound data in a particular frequency band.
And S40, determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values.
In this embodiment, after the feature value of the sound parameter is obtained, since the sound parameter has a mapping relationship with the subjective evaluation word, the evaluation data of the sound data on the subjective evaluation word may be calculated based on the mapping relationship. Illustratively, a subjective term is dynamics (A) and the matching sound parameter is low frequencyExtension parameter (LFX) and low frequency harmonic distortion parameter (THD)L). In determining LFX and THDLAfter the value of (a), then a ═ f (LFX, THD)L) Wherein f is used to represent A and (LFX, THD)L) The mapping relationship between them.
In steps S10-S40, the evaluation object (sound data) and the evaluation attribute (subjective evaluation word) are obtained by acquiring the sound data and the subjective evaluation word. And acquiring sound parameters matched with the subjective evaluation words, wherein the sound parameters are objective parameters, and the matching relation between the sound parameters and the subjective evaluation words is determined based on a preset regression model. And determining the characteristic value of the sound parameter according to the sound data, wherein the characteristic value of the sound parameter is extracted from the sound data, and the obtained characteristic value has uniqueness. And determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values to obtain the finally required evaluation data, wherein the evaluation data can replace the evaluation data obtained by a subjective evaluation method to a certain extent, so that the evaluation time of the evaluation data is reduced, and the reproducibility and the accuracy of the evaluation data are improved.
Optionally, if the subjective evaluation word is a strength, the sound parameters matched with the subjective evaluation word include a low-frequency extension parameter and a low-frequency harmonic distortion parameter;
step S40, namely, the determining the evaluation data of the sound data on the subjective evaluation word according to the feature value includes:
processing the characteristic value through a force formula to generate force evaluation data, wherein the force formula comprises:
A＝a+b*LFX+c*THDL；
wherein, A is dynamics evaluation data, LFX is the characteristic value of low-frequency extension parameter, THDLAnd a, b and c are constant coefficients which are characteristic values of the low-frequency harmonic distortion parameters.
In this embodiment, historical sound parameters and subjective evaluation data are analyzed based on a regression model, and a mapping relationship between two sound parameters, namely a low-frequency extension parameter and a low-frequency harmonic distortion parameter, and the strength is determined from a plurality of sound parameters. In an example, by solving the regression model, specific values of each constant coefficient in the force formula can be obtained, specifically: a is 1.678, b is-0.119, and c is-0.009.
Optionally, in step S30, the determining the feature value of the sound parameter according to the sound data includes:
calculating the characteristic value of the low-frequency extension parameter by using a low-frequency extension formula, wherein the low-frequency extension formula is as follows:
wherein, yREF1The first reference amplitude is the sum of the average amplitude of the audio in the first reference frequency band and a preset specified value; the frequency range of the first reference frequency band is 300Hz-10 kHz;
indicating that in the frequency band below 300Hz, the first amplitude is below yREF1A frequency of value 6 dB;
calculating the characteristic value of the low-frequency harmonic distortion parameter through a low-frequency harmonic distortion formula, wherein the low-frequency harmonic distortion formula is as follows:
wherein N is50-300HzIs the total number of frequency points measured in the frequency range of 50Hz to 300Hz, fn1Is n th1Frequency value corresponding to each frequency point, n1Is an integer, and 0 < n1≤N50-300Hz；THD(fn1) Is a and fn1Characteristic values of the corresponding harmonic distortion parameters.
In the present embodiment, the first and second electrodes are,
meaning that the amplitude is lower than yREF1A value of 6dB, less than and closest to the frequency of 300 Hz.That is to say that the position of the first electrode,
is in the frequency band below 300Hz, all amplitudes are below yREF1Of the frequencies having a value of 6dB, the frequency having the highest frequency value. When the characteristic value of the low-frequency extension parameter is calculated, the linear mapping relation between the subjective data and the objective data is conveniently constructed in the regression model by taking the logarithm. Due to the fact that the low frequency of the interior sound equalizer is improved, 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, typically +10 dB. The first reference frequency band may be selected to be generally 300Hz to 10 kHz. When the method is used for evaluating the sound field in the vehicle, the first reference frequency band can be set to be 300 Hz-3 kHz.
Here, the characteristic value of the low frequency harmonic distortion parameter may be an average of all Total Harmonic Distortion (THD) in a low frequency band (50-300 Hz). The value of n can be set as desired, and illustratively, can be sampled every 10 Hz. Total Harmonic Distortion (THD) refers to the extra harmonic component of the output signal that is caused by the non-linear element to be more than the input signal when the audio signal source passes through the power amplifier.
Optionally, if the subjective evaluation word is brightness, the sound parameters matched with the subjective evaluation word include a high-frequency extension parameter and a high-frequency quality parameter;
step S40, namely, the determining the evaluation data of the sound data on the subjective evaluation word according to the feature value includes:
processing the characteristic values through a brightness formula to generate brightness evaluation data, wherein the brightness formula comprises:
B＝d+e*HFQ+f*HFX；
wherein, B is brightness evaluation data, HFX is a characteristic value of a high-frequency extension parameter, HFQ is a characteristic value of a high-frequency quality parameter, and d, e and f are constant coefficients.
In this embodiment, the matching relationship between the brightness of the subjective evaluation word and the two sound parameters, i.e., the high-frequency extension parameter and the high-frequency quality parameter, is determined based on the analysis result of the regression model. In an example, by solving the regression model, specific values of each constant coefficient in the brightness formula can be obtained, specifically: d-6.378, e-0.359, and f-0.081.
Optionally, the determining the feature value of the sound parameter according to the sound data includes:
calculating the characteristic value of the high-frequency extension parameter through a high-frequency extension formula, wherein the high-frequency extension formula is as follows:
wherein, yREF2Is a second reference amplitude, the second reference amplitude is the average amplitude of the audio frequency in a second reference frequency band, the second reference frequency band is 300-3 kHz,
indicating that in the frequency band above 5kHz, the first is lower than yREF2A frequency of 6 dB;
calculating the characteristic value of the high-frequency quality parameter through a high-frequency quality formula, wherein the high-frequency quality formula is as follows:
wherein N is1Is the total number of 1/20 octave bands between 5kHz and the highest frequency defined by HFX,
denotes the n-th2Average amplitude, n, corresponding to each octave band2Is an integer, and 0 < n2≤N1。
In the present embodiment, the first and second electrodes are,
meaning that the amplitude is lower than yREF2A value of 6dB, greater than and closest to the frequency of 5 kHz. That is to say that the position of the first electrode,
in the frequency band above 5kHz, all amplitudes are below yREF2Of the frequencies having a value of 6dB, the frequency having the lowest frequency value. When the characteristic value of the high frequency extension parameter (HFX) is calculated, the linear mapping relation between the subjective data and the objective data is conveniently constructed in the regression model by taking the logarithm. HFX theoretically correlates positively with subjective evaluation scores.
The highest frequency defined by HFX may be the highest sampling frequency of the sound data. For example, if the highest sampling frequency of the audio data is 30kHz, the highest frequency defined by HFX is 30 kHz. The high frequency quality parameter (HFQ) is used to quantify the amplitude response deviation of the treble region between 5kHz to the high frequency cutoff frequency. Here, the HFX and HFQ parameters are related to the brightness of the subjective evaluation.
Optionally, if the subjective evaluation word is a human voice, the voice parameters matched with the subjective evaluation word include a narrow-band deviation, a frequency spectrum flatness and an intermediate frequency harmonic distortion parameter;
step S40, namely, the determining the evaluation data of the sound data on the subjective evaluation word according to the feature value includes:
processing the characteristic value through a voice formula to generate voice evaluation data, wherein the voice formula comprises:
C＝g+h*NBD+i*SPF+j*THDM；
wherein C is the evaluation data of human voice, NBD is the characteristic value of narrow-band deviation, SPF is the characteristic value of spectral flatness, THDMAnd g, h, i and j are constant coefficients.
In this embodiment, the historical sound parameters and the subjective evaluation data are analyzed based on the regression model, and a mapping relationship between the three sound parameters, namely the narrowband deviation, the spectral flatness, and the medium-frequency harmonic distortion parameter, and the human voice is determined from the multiple sound parameters. In an example, by solving the regression model, specific values of each constant coefficient in the force formula can be obtained, specifically: 14.822, h-2.113, 1.523, and 3.824.
Optionally, in step S30, the determining the feature value of the sound parameter according to the sound data includes:
calculating a characteristic value of the narrow-band deviation by a narrow-band deviation formula, wherein the narrow-band deviation formula comprises:
wherein N is2Is the total number of 1/2 octave bands between 100Hz and 12kHz,
is the n-th3Average amplitude value, y, in 1/2 octave bandbIs the n-th3Band b within 1/2 octave bandfAmplitude value of n3Is an integer, and N is more than 0 and less than or equal to N2；
Calculating a characteristic value of the spectral flatness through a spectral flatness formula, wherein the spectral flatness formula is as follows:
wherein s iskThe value of a signal amplitude spectrum on a frequency point K is shown, wherein the K is the upper limit frequency of a preset frequency range, and the preset frequency range comprises 300Hz-20 kHz;
calculating the characteristic value of the intermediate frequency harmonic distortion parameter through an intermediate frequency harmonic distortion formula, wherein the intermediate frequency harmonic distortion formula is as follows:
wherein N is300-3000HzIs the total number of frequency points measured in the 300Hz to 3000Hz frequency band, fn4Is n th4Frequency value corresponding to each frequency point, THD (f)n4) Is a and fn4Characteristic values of the corresponding harmonic distortion parameters.
In this embodiment, in calculating the eigenvalues of the narrow band bias, the mean absolute bias in each 1/2 octave band is based on 10 equally logarithmically spaced sampled amplitude data points within the octave band. Higher values of NBD indicate greater amplitude deviation in the narrow band. Here, the feature value of NBD is inversely related to the subjective evaluation score.
Spectral flatness (SPF), i.e. the ratio of the geometric mean to the arithmetic mean of the power spectrum of a signal. The ratio is between 0 and 1, and the calculation result is 1 for white noise; for a pure tone signal, the result is 0. The closer the SPF is to 1, the higher the flatness is indicated. The logarithm is taken to enlarge the value range, and the result after the logarithm is taken is larger (closer to 0), so that the flatness is higher. The preset frequency range of the SPF is 300Hz-20 kHz. The SPF is positively correlated with the subjective score, and its feature value is correlated with the voice of the person being evaluated (i.e., the pronunciation of the person is different and the SPF is also different).
The characteristic value of the intermediate frequency harmonic distortion parameter may be an average of all Total Harmonic Distortion (THD) in the intermediate frequency band (300Hz to 3000 Hz). n is4The value of (c) can be set as desired.
Optionally, if the subjective evaluation word is the overall tone quality, the sound parameters matched with the subjective evaluation word include a low-frequency extension parameter, a low-frequency harmonic distortion parameter, a high-frequency extension parameter, a high-frequency quality parameter, a narrow-band deviation, a frequency spectrum flatness, and a medium-frequency harmonic distortion parameter;
the determining of the evaluation data of the sound data on the subjective evaluation words according to the feature values includes:
processing the characteristic value through an integral tone quality formula to generate integral tone quality evaluation data, wherein the integral tone quality formula comprises:
S＝k+l*LFX+m*THDL+n*HFQ+o*HFX+p*NBD+q*SPF+r*THDM；
wherein S is the overall sound quality evaluation data,
LFX is a characteristic value of the low frequency extension parameter,
THDLis the eigenvalue of the low frequency harmonic distortion parameter,
HFX is a characteristic value of the high frequency extension parameter,
HFQ is a characteristic value of a high frequency quality parameter,
NBD is a characteristic value of the narrow-band deviation,
SPF is a characteristic value of spectral flatness,
THDMis a characteristic value of the intermediate frequency harmonic distortion parameter,
k. l, m, n, o, p, q, r are constant coefficients.
In this embodiment, the historical sound parameters and the subjective evaluation data are analyzed based on the regression model, and the sound parameters having mapping relationships with the strength, the brightness, and the human voice are respectively determined. The overall timbre evaluation data may be a weighted average of the evaluation data of the three subjective evaluation words. In one example, the weighting coefficients of the evaluation data of the three subjective evaluation words are equal, and at this time, the constant coefficients in the overall sound quality formula are respectively: k is 7.626 (i.e., k is 1/3(a + d + g), and so on), l is-0.04 (i.e., l is 1/3b), m is-0.003, n is-0.12, o is 0.027, p is-0.704, q is 0.508, and r is-1.275.
The calculation process of the feature values of each sound parameter may refer to the calculation process of the previous embodiment, and is not described herein again.
In the regression model, in order to determine the relative importance of the sound parameters, the regression coefficients were normalized and sorted by absolute value, and the results are shown in table 1.
TABLE 1 normalization coefficient for each acoustic parameter in the regression model
| Predicted variables | Normalized coefficient |
| SPF | 0.245 |
| HFX | 0.212 |
| NBD | -0.165 |
| THDM | -0.121 |
| HFQ | -0.106 |
| LFX | -0.006 |
In table 1, SPF and HFX are most important for the dependent variable, with positive signs indicating positive correlation with the dependent variable and negative signs indicating negative correlation with the dependent variable. Thus, SPF, HFX and THD can be increasedMThe method (2) improves subjective evaluation scores. As shown in fig. 2, fig. 2 is 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 in the present embodiment.
And (5) verifying the regression model by adopting leave-one-out cross verification. And selecting 1 sample from the 6 samples as a test sample, using the other 5 samples as training samples to obtain a regression model, and analyzing the prediction effect of the regression model on the test sample. The selection of test samples is repeated until each sample is used as a test sample. The verification data is shown in table 2.
TABLE 2 verification data of regression models
In table 2, the predicted relative error is substantially within 15%. The average absolute relative error was 9.93%, which was less than 10%. The root mean square error RMSE is 0.81 and less than 0.9, which indicates that the regression model has certain prediction effect.
In one example, a staged sound test (including before tuning and after tuning) is performed on two cars with the same model and the same level of sound hardware, and corresponding sound data is obtained. The evaluation data of the sound data were evaluated by using the subjective evaluation method and the evaluation methods provided in the examples of the present invention, as shown in table 3.
TABLE 3 evaluation data obtained by different evaluation methods
S1Evaluation score obtained for subjective evaluation method, S2The evaluation score obtained by the evaluation method provided by the embodiment of the invention.
In table 3, after tuning, the evaluation scores obtained by the evaluation methods provided in the examples of the present invention all increased and were consistent with the evaluation scores obtained by the subjective evaluation method. The evaluation score of the embodiment of the invention on the aspect of the dynamics is obviously increased, and correspondingly, the evaluation score of the subjective evaluation method on the dynamics is also obviously increased. Therefore, the sound quality evaluation method provided by the embodiment of the invention has high accuracy of evaluation data in terms of strength. In both the brightness and the human voice, the evaluation score obtained in the embodiment of the present invention does not increase as much as the evaluation score of the subjective evaluation method, but does not change negatively. Thus, in general, the evaluation method provided by the embodiment of the invention can be used for evaluating the sound quality before and after tuning of the vehicle.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 3. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile 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 operating system and execution of computer-readable instructions in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external server through a network connection. The computer readable instructions, when executed by a processor, implement a method for acoustic quality assessment.
In one embodiment, a computer device is provided, comprising a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, the processor when executing the computer readable instructions implementing the steps of:
acquiring sound data and subjective evaluation words;
acquiring sound parameters matched with the subjective evaluation words;
determining a characteristic value of the sound parameter according to the sound data;
and determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values.
In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided, the readable storage media provided by the embodiments including non-volatile readable storage media and volatile readable storage media. The readable storage medium has stored thereon computer readable instructions which, when executed by one or more processors, perform the steps of:
acquiring sound data and subjective evaluation words;
acquiring sound parameters matched with the subjective evaluation words;
determining a characteristic value of the sound parameter according to the sound data;
and determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values.
It will be understood by those of ordinary skill in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware associated with computer readable instructions, which can be stored in a non-volatile computer readable storage medium, and when executed, can include processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A sound quality evaluation method is characterized by comprising:
acquiring sound data and subjective evaluation words;
acquiring sound parameters matched with the subjective evaluation words;
determining a characteristic value of the sound parameter according to the sound data;
and determining the evaluation data of the sound data on the subjective evaluation words according to the characteristic values.
2. The sound quality evaluation method according to claim 1, wherein if the subjective evaluation term is dynamics, the sound parameters matched with the subjective evaluation term include a low-frequency extension parameter and a low-frequency harmonic distortion parameter;
the determining of the evaluation data of the sound data on the subjective evaluation words according to the feature values includes:
processing the characteristic value through a force formula to generate force evaluation data, wherein the force formula comprises:
A＝a+b*LFX+c*THDL；
wherein, A is dynamics evaluation data, LFX is the characteristic value of low-frequency extension parameter, THDLAnd a, b and c are constant coefficients which are characteristic values of the low-frequency harmonic distortion parameters.
3. The sound quality evaluation method according to claim 2, wherein the determining the feature value of the sound parameter from the sound data includes:
calculating the characteristic value of the low-frequency extension parameter by using a low-frequency extension formula, wherein the low-frequency extension formula is as follows:
wherein, yREF1The first reference amplitude is the sum of the average amplitude of the audio in the first reference frequency band and a preset specified value; the frequency range of the first reference frequency band is 300Hz-10 kHz;
indicating that in the frequency band below 300Hz, the first amplitude is below yREF1A frequency of value 6 dB;
calculating the characteristic value of the low-frequency harmonic distortion parameter through a low-frequency harmonic distortion formula, wherein the low-frequency harmonic distortion formula is as follows:
wherein N is50-300HzIs the total number of frequency points measured in the frequency range of 50Hz to 300 Hz; f. ofn1Is n th1Frequency value corresponding to each frequency point, n1Is an integer, and 0 < n1≤N50-300Hz；THD(fn1) Is a and fn1Characteristic values of the corresponding harmonic distortion parameters.
4. The acoustic quality evaluation method according to claim 1, wherein if the subjective evaluation word is brightness, the sound parameters matched with the subjective evaluation word include a high-frequency extension parameter and a high-frequency quality parameter;
the determining of the evaluation data of the sound data on the subjective evaluation words according to the feature values includes:
processing the characteristic values through a brightness formula to generate brightness evaluation data, wherein the brightness formula comprises:
B＝d+e*HFQ+f*HFX；
wherein, B is brightness evaluation data; HFX is a characteristic value of the high-frequency extension parameter; HFQ is a characteristic value of the high-frequency quality parameter; d. e and f are constant coefficients.
5. The sound quality evaluation method according to claim 4, wherein the determining the feature value of the sound parameter from the sound data includes:
calculating the characteristic value of the high-frequency extension parameter through a high-frequency extension formula, wherein the high-frequency extension formula is as follows:
wherein, yREF2The second reference amplitude is the average amplitude of audio in a second reference frequency band, and the second reference frequency band is 300-3 kHz;
indicating that in the frequency band above 5kHz, the first amplitude is lower than yREF2A frequency of value 6 dB;
calculating the characteristic value of the high-frequency quality parameter through a high-frequency quality formula, wherein the high-frequency quality formula is as follows:
wherein N is1Is the total number of 1/20 octave bands between 5kHz and the highest frequency defined by HFX;
denotes the n-th2Average amplitude, n, corresponding to each octave band2Is an integer, and 0 < n2≤N1。
6. The sound quality evaluation method of claim 1, wherein if the subjective evaluation word is human sound, the sound parameters matched with the subjective evaluation word include narrow-band deviation, spectral flatness, and medium-frequency harmonic distortion parameters;
the determining of the evaluation data of the sound data on the subjective evaluation words according to the feature values includes:
processing the characteristic value through a voice formula to generate voice evaluation data, wherein the voice formula comprises:
C＝g+h*NBD+i*SPF+j*THDM；
wherein C is the evaluation data of human voice, NBD is the characteristic value of narrow-band deviation, SPF is the characteristic value of spectral flatness, THDMAnd g, h, i and j are constant coefficients.
7. The sound quality evaluation method according to claim 6, wherein the determining the feature value of the sound parameter from the sound data includes:
calculating a characteristic value of the narrow-band deviation by a narrow-band deviation formula, wherein the narrow-band deviation formula comprises:
wherein N is2Is the total number of 1/2 octave bands between 100Hz and 12kHz,
is the n-th3Average amplitude value, y, in 1/2 octave bandbIs the n-th3Band b within 1/2 octave bandfAmplitude value of n3Is an integer, and 0 < n3≤N2；
Calculating a characteristic value of the spectral flatness through a spectral flatness formula, wherein the spectral flatness formula is as follows:
wherein s iskThe value of a signal amplitude spectrum on a frequency point K is shown, wherein the K is the upper limit frequency of a preset frequency range, and the preset frequency range comprises 300Hz-20 kHz;
calculating the characteristic value of the intermediate frequency harmonic distortion parameter through an intermediate frequency harmonic distortion formula, wherein the intermediate frequency harmonic distortion formula is as follows:
wherein N is300-3000HzIs the total number of frequency points measured in the 300Hz to 3000Hz frequency band, fn4Is n th4Frequency value corresponding to each frequency point, THD (f)n4) Is a and fn4Characteristic values of the corresponding harmonic distortion parameters.
8. The sound quality evaluation method of claim 1, wherein if the subjective evaluation term is overall sound quality, the sound parameters matched with the subjective evaluation term include a low frequency extension parameter, a low frequency harmonic distortion parameter, a high frequency extension parameter, a high frequency quality parameter, a narrow band deviation, a spectral flatness, and a medium frequency harmonic distortion parameter;
the determining of the evaluation data of the sound data on the subjective evaluation words according to the feature values includes:
processing the characteristic value through an integral tone quality formula to generate integral tone quality evaluation data, wherein the integral tone quality formula comprises:
S＝k+l*LFX+m*THDL+n*HFQ+o*HFX+p*NBD+q*SPF+r*THDM；
wherein S is the overall sound quality evaluation data,
LFX is a characteristic value of the low frequency extension parameter,
THDLis the eigenvalue of the low frequency harmonic distortion parameter,
HFX is a characteristic value of the high frequency extension parameter,
HFQ is a characteristic value of a high frequency quality parameter,
NBD is a characteristic value of the narrow-band deviation,
SPF is a characteristic value of spectral flatness,
THDMis a characteristic value of the intermediate frequency harmonic distortion parameter,
k. l, m, n, o, p, q, 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, wherein the processor when executing the computer readable instructions implements the sound quality assessment method according to any one of claims 1 to 8.
10. One or more readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the sound quality assessment method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010745297.XA CN114067841A (en) | 2020-07-29 | 2020-07-29 | Sound quality evaluation method, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010745297.XA CN114067841A (en) | 2020-07-29 | 2020-07-29 | Sound quality evaluation method, computer device, and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114067841A true CN114067841A (en) | 2022-02-18 |
Family
ID=80226792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010745297.XA Pending CN114067841A (en) | 2020-07-29 | 2020-07-29 | Sound quality evaluation method, computer device, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114067841A (en) |
-
2020
- 2020-07-29 CN CN202010745297.XA patent/CN114067841A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8565908B2 (en) | Systems, methods, and apparatus for equalization preference learning | |
| TWI440018B (en) | Loudness measurement with spectral modifications | |
| US20110058685A1 (en) | Method of separating sound signal | |
| JP5542206B2 (en) | Method and system for determining perceptual quality of an audio system | |
| KR20130008609A (en) | Apparatus and method for modifying an input audio signal | |
| CN104919525B (en) | For the method and apparatus for the intelligibility for assessing degeneration voice signal | |
| CN109473091B (en) | Voice sample generation method and device | |
| JP6371516B2 (en) | Acoustic signal processing apparatus and method | |
| CN106663450B (en) | Method and apparatus for evaluating quality of degraded speech signal | |
| US20140272883A1 (en) | Systems, methods, and apparatus for equalization preference learning | |
| JP2011501206A (en) | Method and system for measuring voice comprehension of audio transmission system | |
| US20190285673A1 (en) | Spectral-dynamics of an audio signal | |
| Moro-Velázquez et al. | Voice pathology detection using modulation spectrum-optimized metrics | |
| US20170024495A1 (en) | Method of modeling characteristics of a musical instrument | |
| CN103054586A (en) | Chinese speech automatic audiometric method based on Chinese speech audiometric dynamic word list | |
| CN111737277B (en) | Sample data processing method and device, computer equipment and storage medium | |
| CN111816207B (en) | Sound analysis method, sound analysis system, automobile and storage medium | |
| CN109600697A (en) | The outer playback matter of terminal determines method and device | |
| CN114067841A (en) | Sound quality evaluation method, computer device, and storage medium | |
| CN109360583B (en) | Tone evaluation method and device | |
| Hohls et al. | Psychoacoustic analysis of HVAC noise with equal loudness | |
| USRE48462E1 (en) | Systems, methods, and apparatus for equalization preference learning | |
| CN116343813A (en) | Chinese speech enhancement method | |
| Mu et al. | A timbre matching approach to enhance audio quality of psychoacoustic bass enhancement system | |
| JP2023070158A (en) | Quantifying signal purity by means of machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |