Speech clarity prediction method, computer device and storage medium

Abstract

The present invention discloses a method for predicting speech clarity, a computer device, and a storage medium. The method includes: performing a noise power test on a tire to obtain a noise power spectrum and a scanned road spectrum corresponding to at least one test condition; performing a first-band road noise simulation process on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition; performing a second-band road noise simulation process on the scanned road spectrum corresponding to each test condition to determine second road noise data corresponding to each test condition; performing an in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each test condition to determine a target speech clarity corresponding to each test condition. This method can predict in-vehicle speech clarity in the early stages of vehicle development, can optimize in-vehicle speech clarity in the early stages of vehicle development, can greatly shorten the vehicle development cycle, and has high application value.

Description

Technical Field

[0001] The present invention relates to the field of automobile development, and in particular to a speech clarity prediction method, computer equipment and storage medium. Background Art

[0002] In-vehicle speech clarity is a crucial indicator that impacts the user's driving experience. Optimizing in-vehicle speech clarity is crucial during vehicle development. Determining in-vehicle speech clarity and optimizing it based on this clarity is therefore a crucial step in vehicle development. Existing technologies typically evaluate and determine in-vehicle speech clarity during the prototype phase of a real vehicle. However, determining in-vehicle speech clarity at this stage places it too late in the development cycle, hindering vehicle development cycles. Therefore, predicting in-vehicle speech clarity early in vehicle development to shorten the development cycle is a pressing technical challenge. Summary of the Invention

[0003] The embodiments of the present invention provide a speech clarity prediction method, a computer device, and a storage medium to solve the problem of how to predict the speech clarity in a vehicle in the early stage of vehicle development.

[0004] A speech intelligibility prediction method, comprising:

[0005] Performing a noise power test on a tire to obtain a noise power spectrum corresponding to at least one test condition and obtaining a scanned road spectrum corresponding to at least one test condition;

[0006] Performing a first frequency band road noise simulation process on the noise power spectrum corresponding to each of the test conditions to determine first road noise data corresponding to each of the test conditions;

[0007] Performing a second frequency band road noise simulation process on the scanned road spectrum corresponding to each of the test conditions to determine second road noise data corresponding to each of the test conditions;

[0008] Performing in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each of the test conditions to determine a target speech clarity corresponding to each of the test conditions;

[0009] The first frequency band and the second frequency band at least partially overlap, and an upper frequency limit of the first frequency band is greater than an upper frequency limit of the second frequency band.

[0010] Preferably, performing a noise power test on a tire to obtain a noise power spectrum corresponding to at least one test condition includes:

[0011] Performing a sound pressure level test on the tire in a rotating drum state to obtain a tire noise sound pressure level corresponding to at least one test condition;

[0012] Performing sound power spectrum conversion on the tire noise sound pressure level corresponding to at least one of the test conditions to obtain a noise power spectrum corresponding to at least one of the test conditions.

[0013] Preferably, performing a first frequency band road noise simulation process on the noise power spectrum corresponding to each of the test conditions to determine first road noise data corresponding to each of the test conditions includes:

[0014] A vehicle statistical energy simulation model is used to perform a first frequency band road noise simulation process on the noise power spectrum corresponding to each of the test conditions to determine first road noise data corresponding to each of the test conditions.

[0015] Preferably, performing a second frequency band road noise simulation process on the scanned road spectrum corresponding to each of the test conditions to determine the second road noise data corresponding to each of the test conditions includes:

[0016] A finite element simulation model is used to perform road noise simulation processing in a second frequency band on the scanned road spectrum corresponding to each of the test conditions to determine second road noise data corresponding to each of the test conditions.

[0017] Preferably, performing in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each of the test conditions to determine a target speech clarity corresponding to each of the test conditions includes:

[0018] performing fitting processing on the first road noise data and the second road noise data corresponding to each of the test conditions to determine target road noise data corresponding to each of the test conditions;

[0019] An in-vehicle speech clarity analysis is performed on the target road noise data corresponding to each of the test conditions to determine the target speech clarity corresponding to each of the test conditions.

[0020] Preferably, the first path noise data includes a first frequency bandwidth and a first path noise pressure level corresponding to the first frequency bandwidth; the second path noise data includes a second frequency bandwidth and a second path noise pressure level corresponding to the second frequency bandwidth;

[0021] The fitting process is performed on the first road noise data and the second road noise data corresponding to each test condition to determine the target road noise data corresponding to each test condition, including:

[0022] Determining an overlapping frequency bandwidth corresponding to each test condition based on the first frequency bandwidth and the second frequency bandwidth corresponding to each test condition;

[0023] Fusing the first-channel noise pressure level and the second-channel noise pressure level within the overlapping frequency bandwidth corresponding to each of the test conditions, and determining the third-channel noise pressure level corresponding to the overlapping frequency bandwidth corresponding to each of the test conditions;

[0024] The first-path noise pressure level outside the overlapping frequency bandwidth, the second-path noise pressure level outside the overlapping frequency bandwidth, and the third-path noise pressure level within the overlapping frequency bandwidth corresponding to each of the test conditions are fitted to determine target road noise data corresponding to each of the test conditions.

[0025] Preferably, the target road noise data includes at least one target frequency and a target sound pressure level corresponding to each target frequency;

[0026] The performing in-vehicle speech clarity analysis on the target road noise data corresponding to each of the test conditions to determine the target speech clarity corresponding to each of the test conditions includes:

[0027] determining a sound pressure level calibration value corresponding to the target frequency based on the target sound pressure level corresponding to each target frequency and the preset sound pressure level corresponding to the target frequency;

[0028] determining the bandwidth speech clarity corresponding to each target frequency based on the sound pressure level calibration value corresponding to each target frequency and the preset weight corresponding to the target frequency;

[0029] The bandwidth speech clarity corresponding to at least one of the target frequencies is processed to determine a target speech clarity corresponding to each of the test conditions.

[0030] Preferably, determining the sound pressure level calibration value corresponding to the target frequency based on the preset sound pressure level corresponding to the target frequency and the target sound pressure level corresponding to the target frequency includes:

[0031] determining a sound pressure level difference corresponding to the target frequency based on a preset sound pressure level corresponding to the target frequency and a target sound pressure level corresponding to the target frequency;

[0032] If the sound pressure level difference is not greater than a first sound pressure level threshold, determining the first sound pressure level threshold as the sound pressure level calibration value corresponding to the target frequency;

[0033] If the sound pressure level difference is greater than the first sound pressure level threshold and less than the second sound pressure level threshold, determining the sound pressure level difference as the sound pressure level calibration value corresponding to the target frequency;

[0034] If the sound pressure level difference is not less than a second sound pressure level threshold, the second sound pressure level threshold is determined as the sound pressure level calibration value corresponding to the target frequency.

[0035] Preferably, after determining the target speech intelligibility corresponding to each test condition, the speech intelligibility prediction method further comprises:

[0036] If the target speech clarity is not greater than the first clarity threshold, determining a frequency to be optimized based on the bandwidth speech clarity corresponding to at least one of the target frequencies and the second clarity threshold;

[0037] Based on the frequency to be optimized, design parameters corresponding to the frequency to be optimized in the road noise simulation model are optimized and adjusted.

[0038] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned speech intelligibility prediction method is implemented.

[0039] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned speech intelligibility prediction method.

[0040] The above-mentioned speech clarity prediction method, computer device and storage medium perform first-band road noise simulation processing on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition, and perform second-band road noise simulation processing on the scanned road spectrum corresponding to each test condition to determine second road noise data corresponding to each test condition. These methods can relatively accurately simulate road noise across the entire frequency band to obtain relatively accurate road noise data. The first and second road noise data obtained by simulation are then used to determine the target speech clarity. This method can predict in-vehicle speech clarity in the early stages of vehicle development, allowing developers to optimize in-vehicle speech clarity in the early stages of vehicle development. This can significantly shorten the vehicle development cycle, save development costs, and has high application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. 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 paying any creative labor.

[0042] Figure 1 is a flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0043] Figure 2 is another flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0044] Figure 3 is another flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0045] Figure 4 is another flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0046] Figure 5 is another flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0047] Figure 6 is another flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0048] Figure 7 is another flow chart of a method for predicting speech intelligibility according to an embodiment of the present invention;

[0049] Figure 8 FIG. 1 is a schematic diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0051] The speech clarity prediction method provided by the embodiment of the present invention is used to predict the speech clarity in a vehicle in the early stage of vehicle development, thereby shortening the vehicle development cycle.

[0052] In one embodiment, if Figure 1 As shown, a speech clarity prediction method is provided, which is applied in Figure 8 The computer device in the example is used to illustrate the process, including the following steps:

[0053] S101: Performing a noise power test on a tire to obtain a noise power spectrum corresponding to at least one test condition and a scanning road spectrum corresponding to at least one test condition;

[0054] S102: performing a first frequency band road noise simulation process on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition;

[0055] S103: performing a second frequency band road noise simulation process on the scanned road spectrum corresponding to each test condition to determine second road noise data corresponding to each test condition;

[0056] S104: Performing in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each test condition to determine a target speech clarity corresponding to each test condition.

[0057] The first frequency band and the second frequency band at least partially overlap, and the upper frequency limit of the first frequency band is greater than the upper frequency limit of the second frequency band.

[0058] In this example, both the first and second frequency bands include upper and lower frequency limits. The upper frequency limit of the first frequency band is greater than the upper frequency limit of the second frequency band, and the lower frequency limit of the first frequency band is less than the upper frequency limit of the second frequency band. This means that the first and second frequency bands at least partially overlap. In this example, the frequencies corresponding to the first frequency band are primarily concentrated in the mid- and high-frequency bands, while the frequencies corresponding to the second frequency band are primarily concentrated in the low-frequency band.

[0059] Among them, the noise power test refers to the test process of performing noise testing on the tires and obtaining the sound power spectrum of the noise. The test condition refers to the condition of simulating road conditions and / or vehicle speed to test the tires, that is, the test condition includes road conditions and / or vehicle speed. The noise power spectrum refers to the power spectrum corresponding to the sound power of the noise under different test conditions. The scanned road spectrum refers to the road spectrum obtained by scanning the actual road conditions and / or vehicle speed according to the test condition. In this example, the scanned road spectrum includes road conditions and / or vehicle speed, which are used to constitute at least one test condition.

[0060] As an example, in step S101, a computer device performs a noise power test on a tire using simulated road conditions and / or vehicle speeds corresponding to at least one test condition. Test data corresponding to each test condition is collected using a microphone and transmitted to the computer device. The computer device then calculates and processes the test data to obtain a noise power spectrum corresponding to each test condition. This is used to simulate road noise in a first frequency band, thereby facilitating in-vehicle speech intelligibility prediction during the early stages of vehicle development. The computer device then obtains a road spectrum generated by scanning the road conditions and / or vehicle speeds corresponding to each test condition and identifies this road spectrum as the scanned road spectrum for the corresponding test condition. It is understood that different road conditions and / or vehicle speeds correspond to different test conditions, and the tires are tested under different test conditions. In this example, a preset number of microphones are positioned a certain distance from the tire to obtain test data generated by the tire under at least one test condition. This test data is used to determine the noise power spectrum. The preset number is generally no less than six, and each microphone is positioned at an equal distance from the tire's ground contact point to reduce acquisition errors. The tire pressure of the tire needs to be consistent with the tire pressure of the entire vehicle in order to more accurately simulate the actual vehicle environment. In this example, the noise power spectrum and the scanning road spectrum are obtained under at least one identical test condition, and the variables are controlled to achieve more accurate simulation of each test condition in the first frequency band and the second frequency band respectively. Among them, the first frequency band refers to a frequency band with a larger frequency. The second frequency band refers to a frequency band with a smaller frequency. It is understandable that the first frequency band and the second frequency band at least partially overlap. For example, the upper frequency limit of the first frequency band is set to 6300Hz, the lower frequency limit of the first frequency band is set to 400Hz, the upper frequency limit of the second frequency band is set to 500Hz, and the lower frequency limit of the second frequency band is set to 20Hz. The frequency range corresponding to the first frequency band is [400Hz, 6300Hz], and the frequency range corresponding to the second frequency band is [20Hz, 500Hz]. The frequency range corresponding to the first frequency band and the frequency range corresponding to the second frequency band overlap in [400 Hz, 500 Hz]. This facilitates the subsequent first frequency band road noise simulation and the second frequency band road noise simulation on the overlapping portion, making the simulation of the overlapping portion more accurate.

[0061] The first-frequency-band road noise simulation refers to the process of performing road noise simulation on the noise power spectrum in the first frequency band. The first road noise data refers to the road noise data obtained by performing the road noise simulation in the first frequency band. The first noise data includes at least one frequency in the first frequency band and the road noise corresponding to each frequency.

[0062] As an example, in step S102, the computer device uses the noise power spectrum corresponding to at least one test condition as a stimulus and performs road noise simulation in a first frequency band. Simulated road noise data for the first frequency band corresponding to each test condition is obtained, and the road noise data for the first frequency band is determined as first road noise data. This method can accurately simulate road noise in the first frequency band and obtain relatively precise first road noise data without requiring actual vehicle verification, thereby enabling prediction of target speech intelligibility in the vehicle during the early stages of vehicle development.

[0063] The second-frequency-band road noise simulation refers to the process of performing road noise simulation on the noise power spectrum in the second frequency band. The second road noise data refers to the road noise data obtained by performing the road noise simulation in the second frequency band. The second road noise data includes at least one frequency in the second frequency band and the road noise corresponding to each frequency.

[0064] As an example, in step S103, the computer device uses the scanned road spectrum corresponding to at least one test condition as a stimulus to perform road noise simulation in the second frequency band, obtains simulated road noise data corresponding to the second frequency band for each test condition, and determines the road noise data corresponding to the second frequency band as the second road noise data. This method can accurately simulate road noise in the second frequency band and obtain relatively precise second road noise data without the need for actual vehicle verification, thereby enabling prediction of target speech intelligibility in the vehicle during the early stages of vehicle development.

[0065] The target speech clarity refers to the in-vehicle speech clarity across the entire frequency band, where the entire frequency band refers to the frequency band corresponding to the union of the first frequency band and the second frequency band.

[0066] As an example, in step S104, the computer analyzes and processes the first and second road noise data corresponding to each test condition to obtain the full-band in-vehicle speech clarity corresponding to each test condition. This full-band in-vehicle speech clarity corresponding to that test condition is then determined as the target speech clarity for that test condition. In this example, by processing the simulated first and second road noise data to determine the target speech clarity, this method enables prediction of in-vehicle speech clarity in the early stages of vehicle development, enabling developers to optimize in-vehicle speech clarity early in the development process, significantly shortening the vehicle development cycle.

[0067] In this embodiment, a first-band road noise simulation is performed on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition. A second-band road noise simulation is performed on the scanned road spectrum corresponding to each test condition to determine second road noise data corresponding to each test condition. This allows for relatively accurate road noise simulation across the entire frequency band to obtain relatively accurate road noise data. The simulated first and second road noise data are then used to determine target speech clarity. This method enables prediction of in-vehicle speech clarity in the early stages of vehicle development, allowing developers to optimize in-vehicle speech clarity in the early stages of vehicle development. This significantly shortens the vehicle development cycle, saves development costs, and has high application value.

[0068] In one embodiment, if Figure 2 As shown, step S101, i.e., performing a noise power test on a tire to obtain a noise power spectrum corresponding to at least one test condition, includes:

[0069] S201: Performing a sound pressure level test on a tire in a rotating drum state to obtain a tire noise sound pressure level corresponding to at least one test condition;

[0070] S202: Performing sound power spectrum conversion on the tire noise sound pressure level corresponding to at least one test condition to obtain a noise power spectrum corresponding to the at least one test condition.

[0071] Among them, the drum state refers to the state in which the drum is working. It is understandable that in the drum state, the road conditions and / or vehicle speeds under different test conditions can be simulated to test the tire under the test condition. In this example, the road conditions corresponding to different test conditions are simulated by replacing different drum surfaces with different degrees of roughness. For example, if the road condition in the test condition is a smooth road surface, the drum surface of the drum is a smooth drum surface; if the road condition in the test condition is a rough road surface, the drum surface of the drum is a rough drum surface. The speed of the actual vehicle is simulated by the rotation speed of the drum. Sound pressure level test refers to a test to obtain the sound pressure level of tire noise under test conditions. Tire noise sound pressure level refers to the sound pressure level generated by the tire under test conditions, which is used to characterize the road noise performance of the tire under test conditions.

[0072] As an example, in step S201, the drum is preheated and, after preheating, enters a rotating state. A tire is placed on a test bench corresponding to the drum and subjected to a sound pressure level test under at least one test condition. Each microphone collects K sound pressure level data corresponding to at least one test condition (where K > 0) and transmits the K sound pressure level data corresponding to each test condition to a computer device. The computer device statistically processes the K sound pressure level data corresponding to each test condition transmitted by each microphone to obtain the tire noise sound pressure level for each test condition. The number of microphones is a preset number. In this example, the computer device obtains K sound pressure level data collected by each microphone under each test condition and performs error analysis on each sound pressure level data against preset data to obtain an error analysis result. The C sound pressure level data (where C ≤ K) whose error analysis results meet the preset error conditions are determined as the sound pressure level data to be processed. The computer device then averages the C sound pressure level data to obtain the tire noise sound pressure level corresponding to each microphone under each test condition. This method ensures that the tire noise sound pressure level corresponding to each test condition is relatively accurate and has a small error. The preset data refers to the preset sound pressure level used for error analysis of the K sound pressure level data collected by the microphone. The error analysis result is the difference between the sound pressure level data and the preset data. For example, the preset error condition is that the difference between the sound pressure level data and the preset data satisfies a value within the range of [-0.2dB, 0.2dB]. If the error analysis result is that the difference between the sound pressure level data and the preset data is within the range corresponding to the preset error condition, it is determined that the error analysis result satisfies the preset error condition; otherwise, the error analysis result does not satisfy the preset error condition.

[0073] Among them, sound power spectrum conversion refers to the process of converting the tire noise sound pressure level to obtain the noise power spectrum.

[0074] As an example, in step S202, after obtaining the tire noise sound pressure level of each microphone under at least one test condition, the computer device converts each tire noise sound pressure level to obtain a noise power spectrum corresponding to each test condition. In this example, the noise power spectrum corresponding to each test condition is determined using a sound power calculation formula, where the sound power calculation formula is:

[0075]

[0076] in, is the noise power spectrum corresponding to a certain test condition. The preset number of microphones. For the The tire noise sound pressure level corresponding to each microphone. , is the distance from the microphone to the center of the tire contact patch. and is the air constant.

[0077] In this example, the tire noise sound pressure level corresponding to each microphone under each test condition is processed, so that the noise power spectrum under each test condition can be determined more accurately.

[0078] In this embodiment, by performing a sound pressure level test on the tire in the rotating drum state, the tire noise sound pressure level corresponding to at least one test condition can be obtained more accurately. By performing power spectrum conversion on the more accurate tire noise sound pressure level, the noise power spectrum corresponding to the test condition can be obtained more accurately, thereby realizing the conversion of the tire noise sound pressure level, making it feasible to subsequently perform simulation tests based on the noise power spectrum corresponding to at least one test condition.

[0079] In one embodiment, step S102, i.e., performing a first-band road noise simulation process on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition, includes: using a vehicle statistical energy simulation model to perform a first-band road noise simulation process on the noise power spectrum corresponding to each test condition to determine first road noise data corresponding to each test condition.

[0080] As an example, a computer device obtains a vehicle statistical energy simulation model (SEA model), sets the frequency band corresponding to each test condition as the first frequency band, determines the noise power spectrum corresponding to each test condition as the excitation, obtains the excitation corresponding to each test condition, loads the excitation corresponding to each test condition into the SEA model, performs simulation processing, obtains road noise data for each test condition in the first frequency band, and determines the road noise data in the first frequency band as the first road noise data for the corresponding test condition. For example, the first frequency band can be selected as [400Hz, 6300Hz]. It is understandable that since the first frequency band is primarily concentrated in the mid- and high-frequency bands, the SEA model provides more accurate simulation results when simulating in these bands. Therefore, the SEA model is used to simulate the actual vehicle in the first frequency band. In this example, to simulate the actual vehicle as accurately as possible, the SEA model includes, but is not limited to, subsystems corresponding to the vehicle body, such as the body sheet metal, hard interior panels, soft interior sound-absorbing and insulating materials, the interior response acoustic cavity, and the exterior field acoustic cavity. Among them, each subsystem of the vehicle statistical energy simulation model creates a connection unit for energy transfer; the body sheet metal and hard interior panels are simulated by the board subsystem; the soft interior sound-absorbing and insulating materials are defined according to the thickness distribution and coverage area of ​​the actual vehicle design; the in-vehicle response sound cavity and the external field sound cavity are simulated by the sound cavity unit, the modal density of the in-vehicle response sound cavity is greater than 5, and the external field sound cavity is simulated using a semi-infinite field.

[0081] In this embodiment, a whole-vehicle statistical energy simulation model is used to simulate the actual vehicle, and the noise power spectrum corresponding to each test condition is used to simulate the whole-vehicle statistical energy simulation model in the first frequency band. This can simulate the test condition corresponding to the actual vehicle as much as possible, and obtain first road noise data that is relatively close to the actual vehicle test, so that the target speech clarity predicted based on the first road noise data is more accurate.

[0082] In one embodiment, step S103, i.e., performing a second-frequency-band road noise simulation process on the scanned road spectrum corresponding to each test condition to determine the second road noise data corresponding to each test condition, includes: using a finite element simulation model to perform a second-frequency-band road noise simulation process on the scanned road spectrum corresponding to each test condition to determine the second road noise data corresponding to each test condition.

[0083] As an example, a computer device obtains a finite element simulation model (a full vehicle FEM model), sets the frequency band corresponding to each test condition as the second frequency band, uses a scanned road spectrum obtained by scanning the road condition and / or vehicle speed corresponding to each test condition as the excitation corresponding to each test condition, loads the excitation corresponding to each test condition into the finite element simulation model, performs simulation processing in the second frequency band, obtains road noise data for each test condition in the second frequency band, and determines the road noise data in the second frequency band as the second road noise data for the corresponding test condition. For example, the second frequency band can be selected as [20 Hz, 500 Hz]. As can be understood, since the second frequency band is primarily concentrated in the low frequency band, the finite element simulation model provides more accurate simulation results when simulating in the low frequency band. Therefore, the finite element simulation model is used to simulate the actual vehicle in the second frequency band to obtain more accurate second road noise data.

[0084] In this embodiment, a finite element simulation model with high simulation accuracy in the second frequency band is used for simulation, so that more accurate second noise data can be obtained, so that the target speech clarity predicted based on the second noise data can be more accurate.

[0085] In one embodiment, if Figure 3 As shown, step S104, i.e., performing in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each test condition to determine the target speech clarity corresponding to each test condition, includes:

[0086] S301: performing fitting processing on first road noise data and second road noise data corresponding to each test condition to determine target road noise data corresponding to each test condition;

[0087] S302: Performing in-vehicle speech clarity analysis on target road noise data corresponding to each test condition to determine target speech clarity corresponding to each test condition.

[0088] The target road noise data refers to the road noise data obtained by fitting the first road noise data and the second road noise data.

[0089] As an example, in step S301, the computer device fits the second road noise data corresponding to the second frequency band and the first road noise data corresponding to the first frequency band under each test condition to obtain road noise data corresponding to the full frequency band under each test condition. This road noise data corresponding to the full frequency band is then determined as the target road noise data for the test condition. For example, if the second frequency band is [20 Hz, 500 Hz] and the first frequency band is [400 Hz, 6300 Hz], then the full frequency band is [20 Hz, 6300 Hz].

[0090] Among them, in-vehicle speech clarity analysis refers to the process of analyzing and processing target road noise data to obtain target speech clarity.

[0091] As an example, in step S302, the computer device performs in-vehicle speech clarity analysis on the target road noise data corresponding to each test condition according to a preset in-vehicle speech clarity analysis standard to obtain a target speech clarity corresponding to each test condition. The preset in-vehicle speech clarity analysis standard includes, but is not limited to, the speech clarity standard of the national standard GB / T15485-1995.

[0092] In this embodiment, target road noise data for the entire frequency band is obtained, and based on the target road noise data, the target speech clarity for the entire frequency band is determined. This method does not rely on actual vehicle testing and can predict the target speech clarity of the vehicle under different test conditions in the early stages of development. This saves manpower and material resources, can effectively shorten the development cycle, and has high application value.

[0093] In one embodiment, the first channel noise data includes a first frequency bandwidth and a first channel noise pressure level corresponding to the first frequency bandwidth; the second channel noise data includes a second frequency bandwidth and a second channel noise pressure level corresponding to the second frequency bandwidth.

[0094] The first frequency bandwidth refers to the bandwidth corresponding to the first frequency band. The first road noise pressure level refers to the sound pressure level of the road noise corresponding to each frequency in the first frequency band. The second frequency bandwidth refers to the bandwidth corresponding to the second frequency band. The second road noise pressure level refers to the sound pressure level of the road noise corresponding to each frequency in the second frequency band.

[0095] Understandably, the road noise pressure level is used to characterize the performance of road noise. If the first frequency band is [400Hz, 6300Hz] and the second frequency band is [20Hz, 500Hz], then the frequency range corresponding to [400Hz, 6300Hz] is the first frequency bandwidth, and each frequency in [400Hz, 6300Hz] corresponds to a first road noise pressure level. The frequency range corresponding to [20Hz, 500Hz] is the second frequency bandwidth, and each frequency in [20Hz, 500Hz] corresponds to a second road noise pressure level. This is used to characterize the performance of the road noise data, so as to facilitate subsequent fitting based on the first and second road noise pressure levels to determine the target road noise data.

[0096] In one embodiment, if Figure 4 As shown, step S301, i.e., fitting the first road noise data and the second road noise data corresponding to each test condition to determine the target road noise data corresponding to each test condition, includes:

[0097] S401: Determine the overlapping frequency bandwidth corresponding to each test condition based on the first frequency bandwidth and the second frequency bandwidth corresponding to each test condition;

[0098] S402: Fusing the first-channel noise pressure level and the second-channel noise pressure level within the overlapping frequency bandwidth corresponding to each test condition to determine the third-channel noise pressure level corresponding to the overlapping frequency bandwidth corresponding to each test condition;

[0099] S403: Fitting the first-path noise pressure level outside the overlapping frequency bandwidth, the second-path noise pressure level outside the overlapping frequency bandwidth, and the third-path noise pressure level within the overlapping frequency bandwidth corresponding to each test condition to determine target road noise data corresponding to each test condition.

[0100] The overlapping frequency bandwidth refers to the frequency bandwidth range in which the first frequency bandwidth and the second frequency bandwidth overlap.

[0101] As an example, in step S401, the computer device analyzes the overlapping bandwidth between the first frequency bandwidth and the second frequency bandwidth corresponding to each test condition to obtain the overlapping frequency bandwidth between the first frequency bandwidth and the second frequency bandwidth under each test condition. For example, if the first frequency band corresponding to the first frequency bandwidth is [400 Hz, 6300 Hz], and the second frequency band corresponding to the second frequency bandwidth is [20 Hz, 500 Hz], and the first frequency bandwidth and the second frequency bandwidth overlap in the bandwidth range corresponding to [400 Hz, 500 Hz], then the overlapping frequency bandwidth corresponds to the frequency range of [400 Hz, 500 Hz].

[0102] The third-path noise pressure level refers to the sound pressure level of the road noise corresponding to each frequency in the overlapping frequency bandwidth.

[0103] As an example, in step S402, after obtaining the overlapping frequency bandwidth between the first frequency bandwidth and the second frequency bandwidth, the computer device fuses the first-channel noise pressure level and the second-channel noise pressure level corresponding to the same frequency in the overlapping frequency bandwidth to obtain the third-channel noise pressure level corresponding to the overlapping frequency bandwidth. For each test condition, the above-mentioned step of obtaining the third-channel noise pressure level is performed to obtain the third-channel noise pressure level corresponding to the overlapping frequency bandwidth under each test condition. In this example, under the same test condition, the first-channel noise pressure level and the second-channel noise pressure level corresponding to the same frequency in the overlapping frequency bandwidth are averaged to obtain the third-channel noise pressure level corresponding to the overlapping frequency bandwidth. For example, when the test condition is a rough road surface and / or the vehicle speed is 36 km / h, and the frequency band corresponding to the overlapping frequency bandwidth is [400 Hz, 500 Hz], the first-channel noise pressure level and the second-channel noise pressure level corresponding to each frequency in [400 Hz, 500 Hz] are averaged to obtain the third-channel noise pressure level corresponding to each frequency in [400 Hz, 500 Hz] when the test condition is a rough road surface and / or the vehicle speed is 36 km / h, that is, the third-channel noise pressure level corresponding to the overlapping frequency bandwidth under the corresponding test condition is obtained. In this example, the first-channel noise pressure level and the second-channel noise pressure level within the overlapping frequency bandwidth are fused, fully considering the simulation results in the first frequency band and the simulation results in the second frequency band, so that the obtained third-channel noise pressure level can be more accurate.

[0104] As an example, in step S403, the computer device performs fitting processing on the first-channel noise pressure level outside the overlapping frequency bandwidth, the second-channel noise pressure level outside the overlapping frequency bandwidth, and the third-channel noise pressure level within the overlapping frequency bandwidth corresponding to each test condition to determine target road noise data corresponding to each test condition. For example, the first frequency band is [400Hz, 6300Hz], the second frequency band is [20Hz, 500Hz], and the overlapping frequency bandwidth corresponds to [400Hz, 500Hz]. The first-channel noise pressure level outside the overlapping frequency bandwidth, the second-channel noise pressure level outside the overlapping frequency bandwidth, and the third-channel noise pressure level within the overlapping frequency bandwidth are fitted to obtain target road noise data, which is image data.

[0105] In this embodiment, the overlapping frequency bandwidth between the first frequency bandwidth and the second frequency bandwidth is obtained, and the first-path noise pressure level and the second-path noise pressure level within the overlapping frequency bandwidth are fused, and the simulation results in the first frequency band and the simulation results in the second frequency band are fully considered to make the obtained third-path noise pressure level more accurate, and thus make the target road noise data obtained by fitting the first-path noise pressure level outside the overlapping frequency bandwidth, the second-path noise pressure level outside the overlapping frequency bandwidth, and the third-path noise pressure level within the overlapping frequency bandwidth more accurate.

[0106] In one embodiment, the target road noise data includes at least one target frequency and a target sound pressure level corresponding to each target frequency.

[0107] The target frequency refers to the frequency within the full frequency band. The target sound pressure level refers to the road noise sound pressure level corresponding to each target frequency obtained from the simulation, and is used to characterize road noise performance. In this example, the target frequency can be determined using a 1 / 3 octave band of the full frequency band. For example, for the full frequency band of [200Hz, 6300Hz], the 1 / 3 octave band of the full frequency band is obtained, resulting in the following target frequencies: 200Hz, 250Hz, 315Hz, 400Hz, 500Hz, 630Hz, 800Hz, 1000Hz, 1250Hz, 1600Hz, 2000Hz, 2500Hz, 3150Hz, 4000Hz, 5000Hz, and 6300Hz.

[0108] In one embodiment, if Figure 5 As shown, step S302, i.e., performing in-vehicle speech clarity analysis on the target road noise data corresponding to each test condition to determine the target speech clarity corresponding to each test condition, includes:

[0109] S501: Determine a sound pressure level calibration value corresponding to the target frequency based on the target sound pressure level corresponding to each target frequency and the preset sound pressure level corresponding to the target frequency;

[0110] S502: Determine the bandwidth speech clarity corresponding to each target frequency based on the sound pressure level calibration value corresponding to each target frequency and the preset weight corresponding to the target frequency;

[0111] S503: Process the bandwidth speech clarity corresponding to at least one target frequency to determine the target speech clarity corresponding to each test condition.

[0112] The preset sound pressure level refers to the upper limit of the sound pressure level corresponding to the target frequency, that is, the maximum sound pressure level at the target frequency. The sound pressure level calibration value refers to the value used to calibrate the target sound pressure level.

[0113] As an example, in step S501, the computer device compares the target sound pressure level corresponding to the target frequency with a preset sound pressure level corresponding to the target frequency, obtains a difference comparison result, and determines a sound pressure level calibration value corresponding to the target frequency based on the difference comparison result. Understandably, because the target sound pressure level affects the in-vehicle speech intelligibility corresponding to the target frequency, if there is a significant difference between the target sound pressure level and the preset sound pressure level for a particular target frequency, it indicates that a sound pressure level calibration value is needed to determine the impact of the target sound pressure level corresponding to the target frequency on speech intelligibility, thereby enabling subsequent prediction of a more accurate target speech intelligibility.

[0114] The preset weight refers to the preset weight used to correct the sound pressure level calibration value. The bandwidth speech clarity refers to the in-vehicle speech clarity corresponding to the target frequency.

[0115] As an example, in step S502, the computer device uses the preset weight corresponding to the target frequency to correct the sound pressure level calibration value corresponding to the same target frequency to obtain a more reasonable bandwidth speech clarity corresponding to each target frequency. In this example, the bandwidth speech clarity corresponding to the target frequency is:

[0116]

[0117] in, is the bandwidth speech clarity corresponding to the j-th target frequency. is the preset weight corresponding to the j-th target frequency. is the sound pressure level calibration value corresponding to the jth target frequency. In this example, the computer device performs the above steps of obtaining the bandwidth speech clarity for each test condition to obtain the bandwidth speech clarity corresponding to each target frequency under each test condition. As shown in Table 1, this is a speech clarity prediction table corresponding to the 1 / 3 octave of a certain type of vehicle under a test condition. As can be seen from Table 1, each target frequency corresponds to a preset sound pressure level upper limit (i.e., preset sound pressure level), target sound pressure level, and sound pressure level calibration value. , preset weights and bandwidth for speech clarity.

[0118] Table 1

[0119]

[0120] As an example, in step S503, the computer device processes the bandwidth speech clarity corresponding to at least one target frequency under each test condition to determine the target speech clarity corresponding to each test condition. In this example, the computer device sums the bandwidth speech clarity corresponding to at least one target frequency under each test condition, processes the summation result using a preset correction coefficient, and determines it as the target speech clarity corresponding to each test condition. For example, the preset correction coefficient is in the range of (0, 1). In this example, the preset correction coefficient is Target speech intelligibility AI for:

[0121]

[0122] in, is the jth target frequency, and m is the number of target frequencies. For example, the computer device sums the speech intelligibility for each bandwidth in Table 1 to obtain the target speech intelligibility for the entire frequency band. For example, for the full frequency band of [200 Hz, 6300 Hz], if the target frequency is determined based on 1 / 3 octaves of the full frequency band, then m is 16.

[0123] In this embodiment, under each test condition, a calibrated sound pressure level value corresponding to each target frequency is obtained. Based on the calibrated sound pressure level value corresponding to each target frequency and a preset weight, a more reasonable bandwidth speech intelligibility can be obtained. Furthermore, based on the bandwidth speech intelligibility, the target speech intelligibility corresponding to each test condition can be more accurately predicted. This method can determine the target speech intelligibility for each test condition without requiring actual vehicle testing. This method can predict the target speech intelligibility for each test condition in the early stages of vehicle development, facilitating in-vehicle speech intelligibility optimization based on the target speech intelligibility for each test condition, thus possessing high application value.

[0124] In one embodiment, if Figure 6 As shown, step S501, based on the preset sound pressure level corresponding to the target frequency and the target sound pressure level corresponding to the target frequency, determines the sound pressure level calibration value corresponding to the target frequency, including:

[0125] S601: Determine a sound pressure level difference corresponding to the target frequency based on a preset sound pressure level corresponding to the target frequency and a target sound pressure level corresponding to the target frequency;

[0126] S602: If the sound pressure level difference is not greater than the first sound pressure level threshold, determining the first sound pressure level threshold as the sound pressure level calibration value corresponding to the target frequency;

[0127] S603: If the sound pressure level difference is greater than the first sound pressure level threshold and less than the second sound pressure level threshold, determine the sound pressure level difference as the sound pressure level calibration value corresponding to the target frequency;

[0128] S604: If the sound pressure level difference is not less than the second sound pressure level threshold, the second sound pressure level threshold is determined as the sound pressure level calibration value corresponding to the target frequency.

[0129] The sound pressure level difference refers to the difference between the preset sound pressure level and the target sound pressure level.

[0130] As an example, in step S601, the computer device calculates the difference between the preset sound pressure level H and the target sound pressure level SPL corresponding to each target frequency under at least one test condition, and obtains the sound pressure level difference (H) corresponding to each target frequency under the corresponding test condition. SPL).

[0131] Among them, the first sound pressure level threshold and the second sound pressure level threshold refer to preset thresholds for judging the size of the sound pressure level difference. The first sound pressure level threshold and the second sound pressure level threshold are both non-negative numbers. The first sound pressure level threshold is less than the second sound pressure level threshold. The first sound pressure level threshold is the lower limit of the sound pressure level difference, for example, it can be set to 0, and the second sound pressure level threshold is the upper limit of the sound pressure level difference, for example, it can be set to 30.

[0132] As an example, in step S602, the computer device determines the sound pressure level difference (H SPL) is not greater than the first sound pressure level threshold, the first sound pressure level threshold is determined as the sound pressure level calibration value corresponding to the target frequency. In this example, the first sound pressure level threshold is set to 0. If the sound pressure level difference (H If the sound pressure level difference (SPL) is not greater than 0, it indicates that the sound pressure level difference is outside the lower limit of the sound pressure level difference and the error is large. It cannot be used to determine the broadband speech intelligibility. The first sound pressure level threshold of 0 is determined as the sound pressure level calibration value corresponding to the target frequency to reduce the error in the subsequent determination of the broadband speech intelligibility.

[0133] As an example, in step S603, when the computer device determines that the sound pressure level difference is greater than the first sound pressure level threshold and less than the second sound pressure level threshold, the computer device determines the sound pressure level difference as the sound pressure level calibration value corresponding to the target frequency. In this example, the first sound pressure level threshold is set to 0 and the second sound pressure level threshold is set to 30. If the sound pressure level difference (H SPL) is in the range of (0, 30], then the sound pressure level difference (H SPL) is determined as the sound pressure level calibration value corresponding to the target frequency. It can be understood that if the sound pressure level difference (H) between the target sound pressure level SPL and the preset sound pressure level H SPL), within the range of the first sound pressure level threshold and the second sound pressure level threshold, it indicates that the sound pressure level difference is normal. The sound pressure level difference (H SPL) is used as the sound pressure level calibration value, so that the bandwidth speech clarity corresponding to the target frequency can be determined more accurately based on the sound pressure calibration value.

[0134] As an example, in step S604, the computer device determines the sound pressure level difference (H SPL) is not less than the second sound pressure level threshold, the second sound pressure level threshold is determined as the sound pressure level calibration value corresponding to the target frequency. In this example, the second sound pressure level threshold is set to 30. When the SPL) is greater than 30, it indicates that the sound pressure level difference is outside the upper limit of the sound pressure level difference and the error is large. The second sound pressure level threshold of 30 is determined as the sound pressure level calibration value to obtain more accurate bandwidth speech clarity.

[0135] In this embodiment, the sound pressure level calibration value corresponding to the target frequency under each test condition can be determined more accurately based on the relationship between the sound pressure level difference corresponding to the target frequency and the first sound pressure level threshold and the second sound pressure level threshold under each test condition.

[0136] In one embodiment, if Figure 7 As shown, after step S101, that is, after determining the target speech intelligibility corresponding to each test condition, the speech intelligibility prediction method further includes:

[0137] S701: If the target speech clarity is not greater than the first clarity threshold, determining a frequency to be optimized based on the bandwidth speech clarity corresponding to at least one target frequency and the second clarity threshold;

[0138] S702: Based on the frequency to be optimized, optimizing and adjusting the design parameters corresponding to the frequency to be optimized in the road noise simulation model.

[0139] The first clarity threshold is a preset threshold used to determine whether target speech clarity needs to be optimized. The second clarity threshold is a threshold used to determine whether bandwidth speech clarity needs to be optimized. The frequency to be optimized is the target frequency corresponding to bandwidth speech clarity that is no greater than the second clarity threshold.

[0140] As an example, in step S701, the computer device determines whether the target speech clarity is greater than a preset first clarity threshold. If the target speech clarity is greater than the preset first clarity threshold, the target speech clarity is determined to meet the vehicle development requirements. If the target speech clarity is not greater than the preset first clarity threshold, the target speech clarity is determined to not meet the vehicle development requirements. If it is determined that the target speech clarity does not meet the vehicle development requirements, it is further determined whether each bandwidth speech clarity corresponding to the target speech clarity is greater than a preset second clarity threshold. If each bandwidth speech clarity is greater than the preset second clarity threshold, the bandwidth speech clarity is determined to meet the vehicle development requirements. If there is a bandwidth speech clarity that is not greater than the preset second clarity threshold, the bandwidth speech clarity is determined to not meet the vehicle development requirements, and the target frequency corresponding to the bandwidth speech clarity that does not meet the vehicle development requirements is determined as the frequency to be optimized.

[0141] As an example, in step S702, after the computer device determines the frequency to be optimized, it selects at least one design parameter in the vehicle statistical energy simulation model and / or the finite element simulation model that affects the target speech clarity corresponding to the frequency to be optimized, optimizes and adjusts the at least one design parameter, and obtains an optimized vehicle statistical energy simulation model and / or optimized finite element simulation model. For example, if the engine in the vehicle statistical energy simulation model and the finite element simulation model affects the target speech clarity at the frequency to be optimized, it is necessary to optimize and adjust the design parameters of the engine in the vehicle statistical energy simulation model and the finite element simulation model. In this example, based on the optimized vehicle statistical energy simulation model and / or the optimized finite element simulation model, the computer device continues to execute steps S101 to S103, and after obtaining the target speech clarity, it continues to determine whether the target speech clarity is greater than a first clarity threshold. If the target speech clarity is not greater than the first clarity threshold, steps S701 to S702 are continued until the target speech clarity is greater than the first clarity threshold, completing the optimization of the target speech clarity and achieving a high level of vehicle NVH performance.

[0142] In this embodiment, after predicting the target speech clarity for each test condition, the design parameters corresponding to the target frequency in the road noise simulation model are optimized based on whether the target speech clarity and the bandwidth speech clarity corresponding to each target frequency exceed the corresponding clarity threshold. This achieves more precise optimization of the target speech clarity, effectively improving the vehicle's road noise performance. This method eliminates the need for actual vehicle testing and optimization, and can achieve the optimized target speech clarity in the early stages of vehicle development, shortening the vehicle development cycle and saving manpower and material resources, thus possessing high application value.

[0143] It should be understood that the size of the serial numbers of the steps in the above embodiments does not 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 embodiments of the present invention.

[0144] In one embodiment, a computer device is provided, whose internal structure diagram can be as follows: Figure 8As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data used or generated during the execution of the speech clarity prediction method. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a speech clarity prediction method is implemented.

[0145] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the speech intelligibility prediction method in the above embodiment is implemented, for example Figure 1 S101-S104 shown, or Figures 2 to 7 To avoid repetition, it will not be described here.

[0146] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for predicting speech intelligibility in the above embodiment is implemented. For example, Figure 1 S101-S104 shown, or Figures 2 to 7 The computer readable storage medium may be non-volatile or volatile.

[0147] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can 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 DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0148] 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 they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for predicting speech intelligibility, characterized in that: include: Performing a noise power test on a tire to obtain a noise power spectrum corresponding to at least one test condition and obtaining a scanned road spectrum corresponding to at least one test condition; Performing a first frequency band road noise simulation process on the noise power spectrum corresponding to each of the test conditions to determine first road noise data corresponding to each of the test conditions; Performing a second frequency band road noise simulation process on the scanned road spectrum corresponding to each of the test conditions to determine second road noise data corresponding to each of the test conditions; Performing in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each of the test conditions to determine a target speech clarity corresponding to each of the test conditions; The first frequency band and the second frequency band at least partially overlap, and an upper frequency limit of the first frequency band is greater than an upper frequency limit of the second frequency band.

2. The speech intelligibility prediction method according to claim 1, wherein The performing of a noise power test on a tire to obtain a noise power spectrum corresponding to at least one test condition includes: Performing a sound pressure level test on the tire in a rotating drum state to obtain a tire noise sound pressure level corresponding to at least one test condition; Performing sound power spectrum conversion on the tire noise sound pressure level corresponding to at least one of the test conditions to obtain a noise power spectrum corresponding to at least one of the test conditions.

3. The speech intelligibility prediction method according to claim 1, wherein The performing a first frequency band road noise simulation process on the noise power spectrum corresponding to each of the test conditions to determine first road noise data corresponding to each of the test conditions includes: A vehicle statistical energy simulation model is used to perform a first frequency band road noise simulation process on the noise power spectrum corresponding to each of the test conditions to determine first road noise data corresponding to each of the test conditions.

4. The method for predicting speech intelligibility according to claim 1, wherein: The performing a second frequency band road noise simulation process on the scanned road spectrum corresponding to each of the test conditions to determine second road noise data corresponding to each of the test conditions includes: A finite element simulation model is used to perform road noise simulation processing in a second frequency band on the scanned road spectrum corresponding to each of the test conditions to determine second road noise data corresponding to each of the test conditions.

5. The speech intelligibility prediction method according to claim 1, wherein: The performing in-vehicle speech clarity analysis on the first road noise data and the second road noise data corresponding to each of the test conditions to determine the target speech clarity corresponding to each of the test conditions includes: performing fitting processing on the first road noise data and the second road noise data corresponding to each of the test conditions to determine target road noise data corresponding to each of the test conditions; An in-vehicle speech clarity analysis is performed on the target road noise data corresponding to each of the test conditions to determine the target speech clarity corresponding to each of the test conditions.

6. The method for predicting speech intelligibility according to claim 5, wherein: The first path noise data includes a first frequency bandwidth and a first path noise pressure level corresponding to the first frequency bandwidth; the second path noise data includes a second frequency bandwidth and a second path noise pressure level corresponding to the second frequency bandwidth; The fitting process is performed on the first road noise data and the second road noise data corresponding to each test condition to determine the target road noise data corresponding to each test condition, including: Determining an overlapping frequency bandwidth corresponding to each test condition based on the first frequency bandwidth and the second frequency bandwidth corresponding to each test condition; Fusing the first-channel noise pressure level and the second-channel noise pressure level within the overlapping frequency bandwidth corresponding to each of the test conditions, and determining the third-channel noise pressure level corresponding to the overlapping frequency bandwidth corresponding to each of the test conditions; The first-path noise pressure level outside the overlapping frequency bandwidth, the second-path noise pressure level outside the overlapping frequency bandwidth, and the third-path noise pressure level within the overlapping frequency bandwidth corresponding to each of the test conditions are fitted to determine target road noise data corresponding to each of the test conditions.

7. The method for predicting speech intelligibility according to claim 5, wherein: The target road noise data includes at least one target frequency and a target sound pressure level corresponding to each target frequency; The performing in-vehicle speech clarity analysis on the target road noise data corresponding to each of the test conditions to determine the target speech clarity corresponding to each of the test conditions includes: determining a sound pressure level calibration value corresponding to the target frequency based on the target sound pressure level corresponding to each target frequency and the preset sound pressure level corresponding to the target frequency; determining the bandwidth speech clarity corresponding to each target frequency based on the sound pressure level calibration value corresponding to each target frequency and the preset weight corresponding to the target frequency; The bandwidth speech clarity corresponding to at least one of the target frequencies is processed to determine a target speech clarity corresponding to each of the test conditions.

8. The method for predicting speech intelligibility according to claim 7, wherein: The determining, based on the preset sound pressure level corresponding to the target frequency and the target sound pressure level corresponding to the target frequency, a sound pressure level calibration value corresponding to the target frequency includes: determining a sound pressure level difference corresponding to the target frequency based on a preset sound pressure level corresponding to the target frequency and a target sound pressure level corresponding to the target frequency; If the sound pressure level difference is not greater than a first sound pressure level threshold, determining the first sound pressure level threshold as the sound pressure level calibration value corresponding to the target frequency; If the sound pressure level difference is greater than the first sound pressure level threshold and less than the second sound pressure level threshold, determining the sound pressure level difference as the sound pressure level calibration value corresponding to the target frequency; If the sound pressure level difference is not less than a second sound pressure level threshold, the second sound pressure level threshold is determined as the sound pressure level calibration value corresponding to the target frequency.

9. The method for predicting speech intelligibility according to claim 7, wherein: After determining the target speech intelligibility corresponding to each test condition, the speech intelligibility prediction method further includes: If the target speech clarity is not greater than the first clarity threshold, determining a frequency to be optimized based on the bandwidth speech clarity corresponding to at least one of the target frequencies and the second clarity threshold; Based on the frequency to be optimized, design parameters corresponding to the frequency to be optimized in the road noise simulation model are optimized and adjusted.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the speech intelligibility prediction method according to any one of claims 1 to 9 is implemented.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the speech intelligibility prediction method according to any one of claims 1 to 9 is implemented.

Images