Sound simulation method and system in virtual scene

Abstract

The invention discloses a sound simulation method and system in a virtual scene, and relates to the technical field of sound simulation, and the method mainly comprises three key steps: firstly, obtaining the multi-mode state parameters of each object in a target virtual scene, such as position, speed and the like, and judging whether the object is in a falling state or not; then, if the object is in a falling state, the collision possibility between the object and other objects is evaluated by using a bounding box algorithm, and if collision occurs, corresponding simulation sound data is analyzed by integrating parameters such as material, motion and quality; and finally, according to information such as the distance, the included angle and the moving speed between each collision sound source and the person in each direction, calculating a collision sound attenuation coefficient, comparing a threshold interval, adjusting the volume and gain values of different frequency bands in a targeted manner, and realizing adjustment of simulation sound parameters in each direction so as to improve the reality sense and immersion sense of sound simulation of a virtual scene.

Description

Technical Field

[0001] The present invention relates to the technical field of sound simulation, and in particular to a sound simulation method and system in a virtual scene. Background Art

[0002] With the rapid development of virtual reality, augmented reality, and mixed reality technologies, users are demanding increasingly higher levels of immersion and realism in virtual scenes. Sound, as a crucial element in creating a sense of realism in virtual scenes, directly impacts the user experience within the virtual environment. Therefore, more accurate and realistic sound simulation methods are needed to meet these evolving user demands. Consequently, a method and system for simulating sound in virtual scenes is needed.

[0003] Existing technologies, such as the invention application patent with publication number CN104134226B, disclose a method, apparatus, and client device for simulating sound in a virtual scene, wherein the method includes: determining whether an object is in a falling state; when the object is in a falling state, detecting whether any area of the physical shape of the object contacts any area of the physical shape of another object in the virtual scene; when any area of the physical shape of the object contacts any area of the physical shape of the other object, retrieving sound data and simulating sound based on the retrieved sound data. The embodiment of the present invention ensures that the image data processing of the falling and contacting objects in the virtual scene matches the processing of the sound data, avoiding the advance or lag of the sound simulation.

[0004] In response to the above solution, this application has found that the above technology has at least the following technical problems: 1. The existing technology may lack comprehensive real-time monitoring and integrated analysis of the multimodal state parameters of the object. It is difficult to accurately identify whether an object has fallen based on only a single or a small number of parameters. For example, if it is judged only by position changes, normal movement may be misjudged as falling, resulting in confusion in subsequent sound simulation. At the same time, when detecting object collisions, either the calculations are complex and inefficient, or the lack of an effective algorithm may lead to misjudgment of the collision, missing the opportunity to simulate the collision sound, or erroneously triggering the sound simulation.

[0005] 2. Existing technologies lack the ability to comprehensively evaluate collision sound characteristics using parameters such as material, motion, and mass. By only considering the material and ignoring the impact of motion speed and object mass on collision sounds, the simulated sound deviates significantly from the actual sound and lacks realism. For example, the sound of a heavy object colliding quickly should be different from the sound of a light object colliding slowly, but existing technologies may not be able to distinguish between them. Furthermore, there is no mechanism to compare the comprehensive evaluation value with the database interval to match the sound data. Sound data may be randomly selected, or the appropriate sound may not be accurately selected based on the actual collision situation, resulting in the sound simulation not fitting the scene.

[0006] 3. Existing technologies lack environmental adaptability and struggle to account for the impact of factors such as the distance, angle, and movement speed between the collision sound source and the person on sound attenuation. They also fail to adjust sound parameters based on the person's actual position and status within the virtual scene, resulting in the same sound heard by people in different locations, undermining the sense of immersion. It's also difficult to properly adjust volume and frequency band gain based on the sound attenuation coefficient, resulting in a consistently flat sound effect. The technology fails to create variations in distance, intensity, and so on based on the relationship between the sound source and the person's position, making it difficult to meet user demands for realistic sound effects. Summary of the Invention

[0007] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a sound simulation method and system in a virtual scene.

[0008] To solve the above technical problems, the present invention adopts the following technical solution: In the first aspect, the present invention provides a sound simulation method in a virtual scene, including: Step 1, judging the falling state of an object: obtaining the multimodal state parameters corresponding to each object in the target virtual scene at the current moment, and then judging whether each object in the target virtual scene is in a falling state.

[0009] Step 2: Detection of object collision properties: If an object in the target virtual scene is in a falling state, evaluate whether the object collides with other objects in the target virtual scene. If it is evaluated that the object in the target virtual scene will collide with another object, analyze the simulated sound data corresponding to the object and the other object in the target virtual scene.

[0010] Step 3: Adjustment of simulated sound parameters: Obtain the personnel information corresponding to each collision sound source and each position in the target virtual scene, and then analyze the adjustment value of the simulated sound parameters corresponding to each collision sound source and each position in the target virtual scene.

[0011] In a second aspect, the present invention provides a sound simulation system in a virtual scene, including: an object falling state judgment module: used to obtain the multimodal state parameters corresponding to each object in the target virtual scene at the current moment, and then judge whether each object in the target virtual scene is in a falling state.

[0012] Object collision property detection module: used to evaluate whether an object in the target virtual scene collides with other objects if the object is in a falling state. If it is evaluated that the object in the target virtual scene will collide with another object, the simulated sound data corresponding to the object and the other object in the target virtual scene are analyzed.

[0013] Simulated sound parameter adjustment module: used to obtain the personnel information corresponding to each collision sound source and each direction in the target virtual scene, and then analyze the simulated sound parameter adjustment value corresponding to each collision sound source and each direction in the target virtual scene.

[0014] The beneficial effects of the present invention are as follows: 1. The embodiment of the present invention can simulate a collision sound that fits the actual situation by accurately judging the falling state of an object, detecting the collision situation, and then analyzing the simulated sound data in combination with the object's material, motion state, and quality parameters. For example, in a virtual game scene, the sound of a metal object falling and colliding and the sound of a wooden object colliding will be significantly different due to the different materials, which makes the sound in the virtual scene more realistic and enhances the user's sense of immersion. At the same time, according to the personnel information corresponding to each collision sound source and each direction, the simulated sound parameters are adjusted, allowing users to feel different sound effects in different positions and directions. Just like in real life, the sound heard will be different depending on the distance from the sound source and the angle. This effect can also be achieved in a virtual scene, making the user feel as if they are in a real environment.

[0015] 2. In an embodiment of the present invention, when an object falls or collides in a virtual scene, the corresponding sound can be quickly simulated to provide the user with immediate auditory feedback. For example, in a virtual building scene, the sound of objects falling and colliding during the construction process can allow users to more intuitively understand the dynamic changes in the scene, thereby enhancing the interactivity of the scene. At the same time, the object state and simulated sound are judged by comprehensively considering the multimodal state parameter information such as the object's position, speed, acceleration, and posture, so that the sound in the virtual scene is closely related to the actual motion state of the object. For example, the sound of an object falling and colliding quickly and the sound of an object falling and colliding slowly will be different due to different motion states, making the virtual scene more realistic and credible.

[0016] 3. In the embodiment of the present invention, when analyzing simulated sound data, the evaluation values of multiple aspects such as material, movement, and quality are comprehensively considered, and a comprehensive collision sound feature evaluation value is obtained through complex calculations, which is then matched with the sound data in the database, thereby improving the accuracy of the sound simulation. Objects of different materials and masses will produce different sounds when they collide, and this comprehensive evaluation can simulate these differences more accurately. At the same time, according to the different threshold ranges of the collision sound attenuation coefficient, the volume and gain values can be flexibly adjusted so that the sound simulation can adapt to different scenarios and user needs. For example, in virtual spaces of different sizes, the propagation and attenuation of sound are different. This adjustment can make the sound effect more reasonable.

[0017] 4. This embodiment of the present invention utilizes a bounding box algorithm for coarse-grained detection when assessing collisions. This method can quickly determine whether collisions are likely between objects, reducing unnecessary computation and improving system performance and efficiency. This detection method can significantly improve system speed, particularly in complex virtual scenes containing numerous objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 work.

[0019] Figure 1 The present invention is a flowchart of the steps for implementing the method.

[0020] Figure 2 This is a schematic diagram of the system module connection of the present invention. DETAILED DESCRIPTION

[0021] 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 the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] The present invention is implemented as follows Figure 1 As shown, a sound simulation method in a virtual scene includes: step 1, judging the falling state of an object: obtaining the multimodal state parameters corresponding to each object in the target virtual scene at the current moment, and then judging whether each object in the target virtual scene is in a falling state.

[0023] In a specific embodiment, the multimodal state parameters corresponding to each object in the target virtual scene are obtained in the following specific acquisition process: In the target virtual scene, a data acquisition module is deployed to monitor the position, velocity, acceleration and posture of each object in real time at a fixed frequency, and the position coordinates of each object in three-dimensional space are obtained through virtual sensor technology. , and the linear velocity corresponding to each object is calculated by the linear velocity formula of position change and angular velocity At the same time, a virtual accelerometer is used to collect acceleration data of each object in each direction. , and use the attitude sensor to obtain the Euler angle of each object to represent the attitude information ,in, Indicates the number corresponding to each object, , m is a positive integer.

[0024] It should be noted that linear velocity is the ratio of displacement to time interval, and angular velocity is the ratio of angular change to time interval.

[0025] In a specific embodiment, the process of determining whether each object in the target virtual scene is in a falling state is as follows: A1. The position coordinates corresponding to each object are , Linear speed , angular velocity , acceleration data and Euler angles to represent attitude information , in the order of position coordinates first, followed by linear velocity, angular velocity, acceleration data, and finally Euler angle attitude information, integrated into a feature vector containing 13 elements and recorded as ,in, Indicates the number corresponding to each element, .

[0026] A2, and the feature vector corresponding to each object , substitute into the calculation formula In the equation, we get the distance between each object and the hyperplane. ,in, is the normal vector of the hyperplane, is the bias term, is the transpose operation.

[0027] It should be noted that the normal vector is a vector perpendicular to the hyperplane. In the target virtual scene, the hyperplane is used to divide whether the object is in a falling state. The normal vector determines the direction of the hyperplane. The bias term is a constant that plays the role of translating the hyperplane in the hyperplane equation. It can be understood as a position adjustment parameter of the hyperplane in the feature space. Adding the bias term to the formula enables the hyperplane to move parallel in space without changing the normal vector. This allows the hyperplane to find a more appropriate division position based on the actual data distribution, so as to more accurately determine whether the object is in a falling state. The transpose operation is a transformation performed on the matrix. The transpose operation can realize the conversion of vector dimensions to meet the requirements of operations such as vector dot product.

[0028] A3. Then compare the distance from each object to the hyperplane with the set distance threshold corresponding to the hyperplane. If the distance from an object to the hyperplane is greater than or equal to the set distance threshold corresponding to the hyperplane, the object is determined to be in a falling state. If the distance from an object to the hyperplane is less than the set distance threshold corresponding to the hyperplane, the object is determined to be not in a falling state. In this way, it is determined whether each object in the target virtual scene is in a falling state.

[0029] Step 2: Detection of object collision properties: If an object in the target virtual scene is in a falling state, evaluate whether the object collides with other objects in the target virtual scene. If it is evaluated that the object in the target virtual scene will collide with another object, analyze the simulated sound data corresponding to the object and the other object in the target virtual scene.

[0030] In a specific embodiment, the evaluation of whether the object in the target virtual scene collides with other objects is performed as follows: the target virtual scene is subjected to a bounding box algorithm and a coarse-grained detection is performed. When the scene is constructed, a minimum rectangular bounding box is constructed for each object, and each side of the minimum rectangular bounding box is parallel to the coordinate axis. When an object in the target virtual scene is in a falling state, the coordinates of the object in the target virtual scene are marked as , the coordinates of other objects are marked as At the same time, the coordinates of the object in the target virtual scene are overlapped with the coordinates of other objects. If the coordinates of the object in the target virtual scene and the coordinates of other objects meet the and , and , and , it is evaluated whether the object in the target virtual scene will collide with the other object.

[0031] In a specific embodiment, the simulated sound data corresponding to the object and the other objects in the target virtual scene are analyzed, and the specific analysis process is as follows: B1. Obtain the material-related parameters, motion state parameters and object mass parameters corresponding to the object and the other objects in the target virtual scene, and then perform material evaluation values, motion evaluation values and object mass evaluation values corresponding to the object and the other objects in the target virtual scene, and then analyze and obtain the comprehensive collision sound feature evaluation values corresponding to the object and the other objects in the target virtual scene.

[0032] B2. Compare the comprehensive collision sound feature evaluation value corresponding to the object and the other object in the target virtual scene with the comprehensive collision sound feature evaluation value interval corresponding to each simulated sound data in the database. If the comprehensive collision sound feature evaluation value corresponding to the object and the other object in the target virtual scene is within the comprehensive collision sound feature evaluation value interval corresponding to a certain simulated sound data in the database, then use the simulated sound data in the database as the simulated sound data corresponding to the object and the other object in the target virtual scene.

[0033] It should be noted that the simulated sound data includes collision sound, friction sound, deformation sound, vibration sound, etc.

[0034] In a specific embodiment, the material evaluation value, motion evaluation value and object quality evaluation value corresponding to the object and the other objects in the target virtual scene are analyzed in the following manner: C1. Normalize the material-related parameters corresponding to the object and the other objects in the target virtual scene, and input them into the material evaluation value analysis model, and finally output the material evaluation value corresponding to the object and the other objects in the target virtual scene, and record it as .

[0035] It should be noted that the material-related parameters include elastic coefficient and friction coefficient. The analysis process of the material evaluation value of the object and the other objects in the target virtual scene is as follows: the elastic coefficient and friction coefficient corresponding to the object and the other objects in the target virtual scene are respectively recorded as and , substitute into the analytical formula , get the material evaluation value corresponding to the object and the other objects in the target virtual scene .

[0036] C2. Normalize the motion state parameters of the object and the other objects in the target virtual scene, and input them into the motion evaluation value analysis model. Finally, output the motion evaluation values of the object and the other objects in the target virtual scene, and record them as .

[0037] It should be noted that the motion state parameters include collision speed and collision angle, and the motion evaluation values corresponding to the object and the other objects in the target virtual scene are obtained according to the analysis process of the material evaluation values corresponding to the object and the other objects in the target virtual scene.

[0038] C3. Normalize the object quality parameters of the object and the other objects in the target virtual scene, and input them into the object quality evaluation value analysis model. Finally, output the object quality evaluation value of the object and the other objects in the target virtual scene, and record it as .

[0039] It should be noted that the object mass parameters include the object mass ratio and the moment of inertia, and the object mass evaluation values corresponding to the object and the other objects in the target virtual scene are obtained by analyzing the material evaluation values corresponding to the object and the other objects in the above-mentioned target virtual scene.

[0040] In a specific embodiment, the analysis obtains a comprehensive collision sound feature evaluation value corresponding to the object and the other objects in the target virtual scene. The specific analysis process is as follows: the material evaluation value, motion evaluation value, and object mass evaluation value corresponding to the object and the other objects in the target virtual scene are substituted into the calculation formula: The comprehensive collision sound feature evaluation value corresponding to the object and the other objects in the target virtual scene is obtained. ,in, 、 、 They are the standard material evaluation value, standard motion evaluation value, and standard object mass evaluation value corresponding to the set object and other objects, 、 、 They are respectively the weight factors corresponding to the material evaluation values of the set object and other objects, the weight factors corresponding to the motion evaluation values, and the weight factors corresponding to the object quality evaluation values.

[0041] It should be noted that 、 、 Both are greater than 0 and less than 1.

[0042] It should also be noted that in terms of materials, the sound characteristic data of various common materials in collision experiments are collected. For example, through professional audio equipment, key parameters such as the frequency, timbre, and loudness of the sound produced when different materials collide are recorded. Taking metal as an example, because its collision sound is crisp and rich in high-frequency components, based on these experimental data, metal is given a relatively high standard material evaluation value; rubber is soft in texture and has a dull collision sound, so a lower standard value is set accordingly. For motion evaluation values, the common motion speed and acceleration ranges of objects in virtual scenes are sorted out. If ordinary objects often fall freely in the scene, their average speed and acceleration are calculated, and the corresponding motion evaluation parameters are set as standard motion evaluation values; if high-speed flying objects are involved, the standard value is determined based on their typical motion parameters. Regarding object mass, a statistical analysis of the mass of objects in the virtual scene is performed, dividing them into different mass intervals. Representative masses are selected from each interval. Based on the actual characteristics of the collision sound produced by objects of this mass, a standard object mass assessment value is determined. A large number of collision scenarios with objects of varying materials, motion states, and mass combinations are simulated in the virtual scene. After each simulation, the calculated comprehensive collision sound characteristic assessment value is compared with the actual recorded collision sound. Using an iterative approach, weighting factors are continuously adjusted until the simulated sound and the actual sound are optimally matched. The theoretical analysis method, based on the principles of sound generation and propagation in physics, analyzes the mechanisms by which material, motion, and object mass influence the various elements of collision sound. For example, material primarily affects the timbre of the sound, while the motion state has a significant impact on the loudness and duration of the sound, and the object's mass affects both the loudness and frequency of the sound. Based on the primary and secondary relationships of these influences, weighting factors for each factor are determined.

[0043] Step 3: Adjustment of simulated sound parameters: Obtain the personnel information corresponding to each collision sound source and each position in the target virtual scene, and then analyze the adjustment value of the simulated sound parameters corresponding to each collision sound source and each position in the target virtual scene.

[0044] In a specific embodiment, the analysis of the simulated sound parameter adjustment values corresponding to each collision sound source in the target virtual scene to each direction, the specific adjustment process is as follows: D1, analyzing the collision sound attenuation coefficient corresponding to each collision sound source in the target virtual scene and each direction, and comparing the collision sound attenuation coefficient corresponding to each collision sound source in the target virtual scene and each direction with the set collision sound attenuation coefficient threshold range.

[0045] D2. If the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are within the first threshold range of the set collision sound attenuation coefficient, the volume is adjusted to 1.3 times the original volume. At the same time, for the 2-5kHz frequency band, the gain value is set to +2.5dB.

[0046] D3. If the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are within the second threshold range of the set collision sound attenuation coefficient, the volume is adjusted to 1.02 times the original volume. At the same time, for the 3-7kHz frequency band, the gain value is set to +0.3dB.

[0047] D4. If the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are within the third threshold range of the set collision sound attenuation coefficient, the volume is adjusted to 0.7 times the original volume. At the same time, for the 5-8kHz frequency band, the gain value is set to -4dB.

[0048] In a specific embodiment, the collision sound attenuation coefficient corresponding to each collision sound source in the target virtual scene and each direction is analyzed. The specific analysis process is as follows: the personnel information corresponding to each collision sound source in the target virtual scene and each direction is obtained. The personnel information includes the personnel distance, angle and moving speed, and is recorded as 、 and , substituted into the calculation formula, The collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are obtained. ,in, Expressed as a natural constant, is expressed as a constant between 0 and 1, Indicates the number corresponding to each collision sound source, , n is a positive integer, Indicates the number corresponding to each direction, , t is a positive integer.

[0049] It should be noted that the distance factor: through logarithmic operation , which can reflect the characteristic that sound attenuates with increasing distance. In reality, the longer the sound propagates, the greater the energy loss and the lower the loudness. The calculation can accurately simulate this law, so that people who are far away from the collision sound source in the virtual scene hear the sound softer, which is in line with the actual auditory experience. Angle factor: through , taking into account the relationship between the sound propagation direction and the position of the person. When the angle between the person and the sound source is different, the received sound intensity will be different. Expressed as a constant between 0 and 1, it can flexibly adjust the degree of influence of the angle on sound attenuation to achieve a more realistic sound directionality simulation. Moving speed factor: through This method reflects the impact of a person's motion on sound reception. If a person moves toward a sound source, their relative velocity changes the intensity of the received sound. Including this factor in the calculation makes the sound attenuation simulation in dynamic scenarios more physically accurate.

[0050] It's also worth noting that by comprehensively considering key factors such as distance, angle, and movement speed, and accurately calculating the collision sound attenuation coefficient, the sound effects in the target virtual scene can be made more realistic than in the real world. In the virtual environment, people in different positions and states can hear realistic variations in sound loudness, enhancing the realism and immersion of the scene. Whether in gaming, virtual training, or simulation scenarios, this provides users with a more realistic auditory experience and improves the overall quality of the virtual experience.

[0051] The present invention is implemented as follows Figure 2 As shown, a sound simulation system in a virtual scene includes: an object falling state judgment module: used to obtain the multimodal state parameters corresponding to each object in the target virtual scene at the current moment, and then judge whether each object in the target virtual scene is in a falling state.

[0052] Object collision property detection module: used to evaluate whether an object in the target virtual scene collides with other objects if the object is in a falling state. If it is evaluated that the object in the target virtual scene will collide with another object, the simulated sound data corresponding to the object and the other object in the target virtual scene are analyzed.

[0053] Simulated sound parameter adjustment module: used to obtain the personnel information corresponding to each collision sound source and each direction in the target virtual scene, and then analyze the simulated sound parameter adjustment value corresponding to each collision sound source and each direction in the target virtual scene.

[0054] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the scope of protection of the present invention.

Claims

1. A sound simulation method in a virtual scene, characterized in that: include: Step 1: Determine the falling state of an object: Obtain the multimodal state parameters corresponding to each object in the target virtual scene at the current moment, and then determine whether each object in the target virtual scene is in a falling state; Step 2: Detecting object collision properties: If an object in the target virtual scene is in a falling state, then evaluating whether the object collides with other objects in the target virtual scene; if it is estimated that the object in the target virtual scene will collide with another object, then analyzing the simulated sound data corresponding to the object and the other object in the target virtual scene; Step 3: Adjustment of simulated sound parameters: Obtain the personnel information corresponding to each collision sound source and each position in the target virtual scene, and then analyze the adjustment value of the simulated sound parameters corresponding to each collision sound source and each position in the target virtual scene.

2. The method for simulating sound in a virtual scene according to claim 1, wherein: The multimodal state parameters corresponding to each object in the target virtual scene are obtained, and the specific acquisition process is as follows: In the target virtual scene, a data acquisition module is deployed to monitor the position, velocity, acceleration and posture of each object in real time at a fixed frequency. Through virtual sensor technology, the position coordinates of each object in three-dimensional space are obtained. , and the linear velocity corresponding to each object is calculated by the linear velocity formula of position change and angular velocity At the same time, a virtual accelerometer is used to collect acceleration data of each object in each direction. , and use the attitude sensor to obtain the Euler angle of each object to represent the attitude information ,in, Indicates the number corresponding to each object, , m is a positive integer.

3. The method for simulating sound in a virtual scene according to claim 2, wherein: The specific process of judging whether each object in the target virtual scene is in a falling state is as follows: A1. The position coordinates of each object , Linear speed , angular velocity , acceleration data and Euler angles to represent attitude information , in the order of position coordinates first, followed by linear velocity, angular velocity, acceleration data, and finally Euler angle attitude information, integrated into a feature vector containing 13 elements and recorded as ,in, Indicates the number corresponding to each element, ; A2, and the feature vector corresponding to each object , substitute into the calculation formula In the equation, we get the distance between each object and the hyperplane. ,in, is the normal vector of the hyperplane, is the bias term, is the transpose operation; A3. Then compare the distance from each object to the hyperplane with the set distance threshold corresponding to the hyperplane. If the distance from an object to the hyperplane is greater than or equal to the set distance threshold corresponding to the hyperplane, the object is determined to be in a falling state. If the distance from an object to the hyperplane is less than the set distance threshold corresponding to the hyperplane, the object is determined to be not in a falling state. In this way, it is determined whether each object in the target virtual scene is in a falling state.

4. The method for simulating sound in a virtual scene according to claim 3, wherein: The specific evaluation process of evaluating whether the object collides with other objects in the target virtual scene is as follows: Run the bounding box algorithm on the target virtual scene and perform coarse-grained detection. When building the scene, a minimum rectangular bounding box is constructed for each object, and its sides are parallel to the coordinate axis. When an object in the target virtual scene is in a falling state, the coordinates of the object in the target virtual scene are marked as , the coordinates of other objects are marked as At the same time, the coordinates of the object in the target virtual scene are overlapped with the coordinates of other objects. If the coordinates of the object in the target virtual scene and the coordinates of some other objects meet the and , and , and , it is evaluated whether the object in the target virtual scene will collide with the other object.

5. The method for simulating sound in a virtual scene according to claim 4, wherein: The specific analysis process of analyzing the simulated sound data corresponding to the object and the other objects in the target virtual scene is as follows: B1. Obtaining material-related parameters, motion state parameters, and object mass parameters corresponding to the object and the other objects in the target virtual scene, thereby analyzing the material evaluation values, motion evaluation values, and object mass evaluation values corresponding to the object and the other objects in the target virtual scene, and further analyzing to obtain a comprehensive collision sound feature evaluation value corresponding to the object and the other objects in the target virtual scene; B2. Compare the comprehensive collision sound feature evaluation value corresponding to the object and the other object in the target virtual scene with the comprehensive collision sound feature evaluation value interval corresponding to each simulated sound data in the database. If the comprehensive collision sound feature evaluation value corresponding to the object and the other object in the target virtual scene is within the comprehensive collision sound feature evaluation value interval corresponding to a certain simulated sound data in the database, then use the simulated sound data in the database as the simulated sound data corresponding to the object and the other object in the target virtual scene.

6. The method for simulating sound in a virtual scene according to claim 5, wherein: The specific analysis process of the material evaluation value, motion evaluation value and object quality evaluation value corresponding to the object and the other objects in the target virtual scene is as follows: C1. Normalize the material parameters of the object and the other objects in the target virtual scene, and input them into the material evaluation value analysis model. Finally, output the material evaluation values of the object and the other objects in the target virtual scene, and record them as ; C2. Normalize the motion state parameters of the object and the other objects in the target virtual scene, and input them into the motion evaluation value analysis model. Finally, output the motion evaluation values of the object and the other objects in the target virtual scene, and record them as ; C3. Normalize the object quality parameters of the object and the other objects in the target virtual scene, and input them into the object quality evaluation value analysis model. Finally, output the object quality evaluation value of the object and the other objects in the target virtual scene, and record it as .

7. The method for simulating sound in a virtual scene according to claim 6, wherein: The analysis obtains the comprehensive collision sound feature evaluation value corresponding to the object and the other objects in the target virtual scene. The specific analysis process is as follows: Substitute the material evaluation value, motion evaluation value, and object quality evaluation value corresponding to the object and the other objects in the target virtual scene into the calculation formula: The comprehensive collision sound feature evaluation value corresponding to the object and the other objects in the target virtual scene is obtained. ,in, 、 、 They are the standard material evaluation value, standard motion evaluation value, and standard object mass evaluation value corresponding to the set object and other objects, 、 、 They are respectively the weight factors corresponding to the material evaluation values of the set object and other objects, the weight factors corresponding to the motion evaluation values, and the weight factors corresponding to the object quality evaluation values.

8. The method for simulating sound in a virtual scene according to claim 1, wherein: The specific adjustment process of analyzing the simulated sound parameter adjustment values corresponding to each direction of each collision sound source in the target virtual scene is as follows: D1. Analyze the collision sound attenuation coefficient corresponding to each collision sound source and each direction in the target virtual scene, and compare the collision sound attenuation coefficient corresponding to each collision sound source and each direction in the target virtual scene with a set collision sound attenuation coefficient threshold range; D2. If the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are within the first threshold range of the set collision sound attenuation coefficient, adjust the volume to 1.3 times the original volume. At the same time, set the gain value to +2.5dB for the 2-5kHz frequency band. D3. If the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are within the second threshold range of the collision sound attenuation coefficient, adjust the volume to 1.02 times the original volume. At the same time, set the gain value to +0.3dB for the 3-7kHz frequency band. D4. If the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are within the third threshold range of the set collision sound attenuation coefficient, the volume is adjusted to 0.7 times the original volume. At the same time, for the 5-8kHz frequency band, the gain value is set to -4dB.

9. The method for simulating sound in a virtual scene according to claim 8, wherein: The specific analysis process of analyzing the collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene is as follows: Obtain the personnel information corresponding to each direction of each collision sound source in the target virtual scene. The personnel information includes personnel distance, angle and moving speed, and is recorded as 、 and , substituted into the calculation formula, The collision sound attenuation coefficients corresponding to each collision sound source and each direction in the target virtual scene are obtained. ,in, Expressed as a natural constant, is expressed as a constant between 0 and 1, Indicates the number corresponding to each collision sound source, , n is a positive integer, Indicates the number corresponding to each direction, , t is a positive integer.

10. A sound simulation system in a virtual scene for executing the sound simulation method in a virtual scene according to any one of claims 1 to 9, characterized in that: include: Object drop state judgment module: used to obtain the multimodal state parameters corresponding to each object in the target virtual scene at the current moment, and then judge whether each object in the target virtual scene is in a drop state; Object collision property detection module: used to evaluate whether an object in the target virtual scene is in a falling state and whether the object collides with other objects in the target virtual scene. If it is estimated that the object in the target virtual scene will collide with another object, the simulated sound data corresponding to the object and the other object in the target virtual scene are analyzed; Simulated sound parameter adjustment module: used to obtain the personnel information corresponding to each collision sound source and each direction in the target virtual scene, and then analyze the simulated sound parameter adjustment value corresponding to each collision sound source and each direction in the target virtual scene.

Images