Sound effect recommendation method, system and equipment based on user preference and scene perception

Abstract

The invention relates to the technical field of Internet information processing, in particular to a sound effect recommendation method, system and device based on user preference and scene awareness, and the method comprises the steps: carrying out the deep analysis of collected user preference data, and constructing a user preference model; when an audio recommendation request of a user side is received, scene perception information matched with a current vehicle position is obtained, and a current scene is determined; matching the current scene with the user preference model to obtain a candidate sound effect set matched with the current scene and the user preference; checking each candidate sound effect data in the candidate sound effect set, screening the candidate sound effect data according to the sound effect data, and generating a recommendation result; and feeding back the recommended sound effect contained in the sound effect recommendation result. According to the scheme, the personalized requirements of the user can be fully considered, and the preferable sound effect recommendation information is provided for the user, so that the satisfaction degree of the user on sound effect recommendation is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet information processing, in particular to a sound effect recommendation method and system based on user preferences and scene perception, a computer device and a storage medium. BACKGROUND

[0002] In order to improve the user's audio experience, the vehicle audio will provide a variety of sound effects for the user to choose, such as sound effects corresponding to different music styles (popular, rock, classical, jazz, etc.), sound effects for different member seating positions (driver, whole vehicle, rear VIP, etc.), sound effects for different immersion modes (stereo, surround sound, 4D, panorama, etc.), and sound effects simulating different environments (such as concert hall, KTV, recording studio, etc.). However, in the prior art, when selecting sound effects, the user often needs to manually try and filter among numerous sound effect options, which is a tedious process and it is difficult to quickly find the sound effect that best meets the user's current needs. In addition, existing sound effect recommendations often do not fully consider the user's individual preferences and the actual use scenario of the device, resulting in a large gap between the recommended sound effects and the user's expectations, and failing to provide the user with the best audio experience. SUMMARY

[0003] To solve the above problems, the present application provides a sound effect recommendation method, system and device based on user preferences and scene perception, to solve the problem of inaccurate sound effect recommendation and tedious user operation in the prior art, and to automatically recommend the most suitable sound effect for the user according to the user's individual preferences and the real-time scenario of the device, thereby improving the user's audio experience.

[0004] To achieve the above-mentioned purpose of the application, the present application adopts the following technical solutions:

[0005] In a first aspect, the present application provides a sound effect recommendation method based on user preferences and scene perception, characterized in that the method comprises:

[0006] Deeply analyzing the collected user preference data to construct a user preference model;

[0007] When receiving an audio recommendation request from the user end, obtaining scene perception information matching the current vehicle position to determine the current scene;

[0008] Matching the current scene with the user preference model to obtain a candidate sound effect set matching the current scene and user preference;

[0009] Viewing each candidate sound effect data in the candidate sound effect set, filtering the candidate sound effect data according to the sound effect data, and generating a recommendation result;

[0010] Feedback of the recommended sound effects contained in the sound effect recommendation result.

[0011] Optionally, the collecting user preference data comprises collecting historical operation data of the user on the in-vehicle audio device and manually input preference setting information of the user.

[0012] The user preference data is obtained according to the historical operation record of the user on the in-vehicle audio device and the manually input preference setting information of the user.

[0013] Optionally, the historical operation record of the matching user information comprises historical operation record of the user on the in-vehicle audio device or the mobile terminal application, including played music file information, playing frequency, stay time, manually selected sound effect mode.

[0014] The behavior data of the user end comprises the like degree of the user end input through the vehicle-mounted interactive interface for different music styles, musical instruments and vocals.

[0015] Optionally, the deep analysis of the collected user preference data and the construction of the user preference model comprise: learning the correlation between the historical operation record of the user on the in-vehicle audio device and the manually input preference setting information of the user through a machine learning algorithm, analyzing the preference degree of the user for different music styles and sound effect types; comprehensively considering the preference degree of the user for different sound effects, the matching degree of the current scene and the sound effect, and the popularity degree of the sound effect in the user group; setting corresponding weights for each factor, and establishing a user preference model.

[0016] Optionally, the obtaining of the scene perception information matching the current vehicle position comprises:

[0017] The current position of the vehicle of the user is located, and all scene perception information covering the current position of the vehicle of the user in a preset area range is displayed, including vehicle position information, navigation information, driving state, ambient light intensity, and ambient noise level information.

[0018] The various scene perception information covering the current position of the vehicle of the user is sequentially sorted according to a predefined rule, and feature comparison is performed to identify the current scene of the vehicle.

[0019] Optionally, the method further comprises: calculating a comprehensive score for each sound effect in the candidate sound effect set according to the factor weight, sequentially sorting the sound effects according to the score from high to low to obtain a sound effect recommendation result; and the sound effect recommendation result comprises at least one optimal sound effect.

[0020] The optimal sound effect is recommended to the user, and the recommendation algorithm is optimized according to the user feedback.

[0021] Optionally, the optimizing the recommendation algorithm according to the user feedback comprises: inputting the user feedback, the current recommendation scene, and user preference data as new training data into an optimization model of the recommendation algorithm for training, and adjusting parameters of the recommendation algorithm model.

[0022] In a second aspect, the present application provides a sound effect recommendation system based on user preference and scene awareness, comprising:

[0023] A model construction module is configured to perform deep analysis on the collected user preference data, and construct a user preference model.

[0024] A determination module is configured to, when receiving an audio recommendation request of a user terminal, acquire scene awareness information matching a current vehicle position, and determine a current scene.

[0025] A matching module is configured to match the current scene with the user preference model, and acquire a candidate sound effect set matched with the current scene and the user preference.

[0026] A screening module is configured to view each candidate sound effect data in the candidate sound effect set, screen the candidate sound effect data according to the sound effect data, and generate a recommendation result.

[0027] A recommendation module is configured to feed back the sound effect recommendation result.

[0028] In a third aspect, the present application provides a computer readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the steps of the sound effect recommendation method based on user preference and scene awareness according to any one of the first aspect.

[0029] In a fourth aspect, the present application provides a computer device comprising a memory and a processor, the memory storing a computer program, the computer program being executed by the processor to make the processor execute the steps of the sound effect recommendation method based on user preference and scene awareness according to any one of the first aspect.

[0030] The beneficial effects of the present application are as follows:

[0031] The application provides a sound effect recommendation method, system and device based on user preference and scene perception, which comprises deep analysis of collected user preference data to build a user preference model; when receiving an audio recommendation request of a user terminal, scene perception information matching a current vehicle position is obtained to determine a current scene; the current scene is matched with the user preference model to obtain a candidate sound effect set matching the current scene and user preference; each candidate sound effect data in the candidate sound effect set is viewed, the candidate sound effect data is screened according to the sound effect data, and a recommendation result is generated; and the recommended sound effect contained in the sound effect recommendation result is fed back. The above scheme can build a precise user preference model by deeply analyzing historical operation records and manual preference settings of a user, can fully consider the individualized needs of the user, recommend sound effects meeting the unique taste of the user, and improve the satisfaction of the user with sound effect recommendation. Moreover, the scene is built by using position information, destination information, weather information and user operation record information, so that the recommended sound effect can be in harmony with the current scene, such as recommending sound effects with fast rhythm and strong bass when driving on a highway for a long distance, recommending soft and soothing sound effects when driving at low speed in light rain, and the like, to provide the user with audio experience more suitable for actual use scenes.

[0032] The application scheme can reasonably recommend sound effects preferred by a user according to a scene where the user is located, thereby helping the user to better select suitable sound effects and avoiding the problem of long time consumption caused by the user's predetermined demand, and improving the operation experience of the user on an operation interface.

[0033] The sound effect recommendation method, system and device based on user preference and scene perception provided by the application can fully consider the individualized needs of a user, provide the user with optimal sound effect recommendation information to improve the satisfaction of the user with sound effect recommendation, and to a certain extent, solve the problem of hysteresis of the recommendation method in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical scheme in the specific embodiments or the prior art, the drawings needed to be used in the specific embodiments or the prior art description will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale.

[0035] Figure 1 A sound effect recommendation method flowchart based on user preference and scene perception provided by the application;

[0036] Figure 2 A sound effect recommendation system structure schematic diagram based on user preference and scene perception provided by the application;

[0037] Figure 3A structural schematic diagram of a computer device provided by the present application is shown. DETAILED DESCRIPTION

[0038] The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0039] In order to specifically understand the technical solutions provided by the present application, the technical solutions of the present application will be described and explained in detail in the following embodiments. Obviously, the embodiments provided by the present application are not limited to the specific details familiar to those skilled in the art. The preferred embodiments of the present application are described in detail as follows, and in addition to these descriptions, the present application can have other embodiments.

[0040] The embodiments of the present application will be described below with reference to the accompanying drawings.

[0041] The present application proposes a sound effect recommendation method based on user preference and scene perception as shown in Figure 1 The method comprises the following steps:

[0042] S101, performing deep analysis on the collected user preference data to construct a user preference model;

[0043] S102, when receiving an audio recommendation request of a user terminal, obtaining scene perception information matching a current vehicle position to determine a current scene;

[0044] S103, matching the current scene with the user preference model to obtain a candidate sound effect set matched with the current scene and the user preference;

[0045] S104, checking each candidate sound effect data in the candidate sound effect set, screening the candidate sound effect data according to the sound effect data, and generating a recommendation result;

[0046] S105, feeding back a recommended sound effect contained in the sound effect recommendation result.

[0047] In the above embodiments, the collection of user preference data in step S101 comprises collecting historical operation data of the user on the whole vehicle audio device and preference setting information manually input by the user;

[0048] According to the historical operation records of the user on the whole vehicle audio device and the preference setting information manually input by the user, the user preference data is obtained.

[0049] In the above embodiments, the matching of the historical operation records of the user information comprises historical operation records of the user on the whole vehicle audio device or a mobile terminal application, including music file information played, playing frequency, stay time, and manually selected sound effect mode;

[0050] The behavior data of the user terminal includes: the user terminal inputs the preference degree of different music styles, musical instruments, and vocals through the vehicle-mounted interactive interface.

[0051] As can be known from the specific steps of step S101, the mobile phone for collecting the user's historical operation records on the whole vehicle audio device and the user's manually input preference setting information provides data support for constructing the user preference model. The audio device continuously records the user's operation behavior information during the user's use process and sends the information to the cloud server. For example, the music file tag information played by the user is recorded, including music style (pop, rock, classical, etc.), singer, album, etc. The user's playing frequency of different music styles is counted, such as playing pop music 20 times and rock music 10 times in a week. The user's staying time on each audio file is recorded to determine the user's preference degree for the audio. For example, the user stays longer on a piece of classical music, which indicates that the user may have a higher preference for classical music. In the setting interface of the audio device, a user preference setting entry is provided. The user can manually input the preference degree of different music styles, musical instruments (such as piano, guitar, violin, etc.), and vocal characteristics (such as high pitch, medium pitch, and low pitch) through a sliding bar, a check box, etc. For example, the user sets the preference degree of piano to “very like” and the preference degree of high-pitched vocals to “general”.

[0052] In the above embodiment, the deep analysis of the collected user preference data to construct the user preference model in step S101 includes: learning the correlation between the user's historical operation records on the whole vehicle audio device and the user's manually input preference setting information through a machine learning algorithm, analyzing the user's preference degree for different music styles and sound effect types; comprehensively considering the user's preference degree for different sound effects, the matching degree of the current scene and the sound effect, and the popularity of the sound effect in the user group; and setting corresponding weights for each factor to establish the user preference model.

[0053] It can be understood that the machine learning algorithm, such as the neural network model in deep learning, is used to analyze the collected user preference data. The user's historical operation records and manual preference settings are used as input data, and the user's preference weights for different music styles and sound effect types are obtained through the training of the neural network. For example, after training, the user's preference weight for the pop music style is 0.8, and the user's preference weight for the classical music style is 0.3. In the above embodiment, the step S102 for obtaining the scene perception information matching the current vehicle position includes:

[0054] Locate the current position of the user's vehicle and display all scene perception information covering the current position of the user's vehicle within the preset area, including vehicle position information, navigation information, driving status, ambient light intensity, and ambient noise level information;

[0055] The scene perception information covering the current position of the user's vehicle is sorted in sequence according to predefined rules, and feature comparison is performed to identify the scene the vehicle is currently in.

[0056] Optionally, during the scene perception phase to match the current vehicle location, the vehicle entertainment system host and vehicle sensors can be used to obtain real-time information about the vehicle's surrounding environment and vehicle motion status, enabling accurate perception of the vehicle's current scene. The accelerometer and gyroscope monitor the vehicle's acceleration changes in real time. When the accelerometer detects large acceleration changes and the gyroscope detects frequent changes in pitch, the system uses pre-set driving mode judgment rules to determine that the vehicle may be in a high-speed driving state. If the acceleration changes detected by the accelerometer are relatively stable and consistent with the vehicle's driving acceleration characteristics, the system determines that the user may be driving at a low speed. The sunlight and rain sensors acquire real-time ambient light intensity and rainfall data. When the light intensity exceeds a preset daytime light threshold, the environment is determined to be daytime; when the light intensity falls below the threshold, the environment is determined to be nighttime. The rain sensor detects rainfall values ​​in different levels, determining whether it is light rain, moderate rain, heavy rain, or torrential rain. Furthermore, the scene judgment is further refined by combining ambient noise data collected by the microphone. For example, if the ambient noise intensity is lower than the quiet environment threshold, it is judged as a quiet driving environment; if the ambient noise intensity is higher than the threshold, it is judged as a relatively noisy driving environment.

[0057] In one embodiment, the above step S103 matches the current scene with the user preference model, and obtaining a candidate sound effect set that matches the current scene and the user preference includes:

[0058] After analyzing the data obtained by the user preference collection module and building a user preference model, we can establish an association model between scenes and sound effects based on the scene information provided by scene perception, match the current scene with user preferences, and screen out a set of candidate sound effects.

[0059] Among them, the association model of the scene and the sound effect can be determined by a large number of experiments and data analysis to determine the appropriate sound effect type in different scenes. For example, the highway driving scene is suitable for sound effects with fast rhythm and low sound enhancement; the low-speed rainy day is suitable for soft sound effects with good noise reduction effect. Then the current perceived scene information is matched with the user preference model to find out the candidate sound effect set that meets the current scene and user preference. For example, the current scene is highway driving, and the user prefers pop music style, then the sound effects with fast rhythm, low sound enhancement and suitable for pop music are selected from the sound effect library to form the candidate sound effect set.

[0060] For example, when the vehicle is in a specific scene, the system first selects the appropriate sound effect mode set (such as “vocal enhancement + moderate low sound enhancement” and “pop music optimization sound effect”) from the scene-sound effect association rule library according to the scene information output by the scene perception module (such as highway sunny cruising). Then, combined with the user's preference probability for each sound effect type output by the user preference model, the sound effect mode set is preliminarily screened, and the sound effect mode with a preference probability greater than 0.6 is retained to form a candidate sound effect set. For example, the user's preference probability for the “vocal enhancement” sound effect is 0.75, and the preference probability for the “pop music optimization sound effect” is 0.68, so both of these two sound effect modes are included in the candidate sound effect set.

[0061] In this embodiment, a multi-objective optimization algorithm (such as NSGA-II algorithm) can be used to consider three objective functions of user preference (preference probability), scene adaptation degree (matching degree with scene-sound effect association rule) and sound effect popularity (usage frequency of this sound effect mode in similar scenes based on cloud big data statistics) to sort the sound effect modes in the candidate sound effect set. The top 3-5 sound effect modes in the ranking are output as the final recommended candidate sound effects.

[0062] Further, the method further comprises: calculating a comprehensive score for each sound effect in the candidate sound effect set according to the factor weight, sequentially ranking the sound effects according to the score from high to low to obtain a sound effect recommendation result; and the sound effect recommendation result comprises at least one optimal sound effect.

[0063] The optimal sound effect is recommended to the user, and the recommendation algorithm is optimized according to the user feedback.

[0064] In the above embodiment, the calculation of the comprehensive score is based on the multi-objective optimization matching result, and the comprehensive score is calculated for each candidate sound effect. The scoring formula is: comprehensive score = a x user preference score + b x scene adaptation score + g x sound effect popularity score, wherein a, b, and g are weight coefficients, and are set to a = 0.5, b = 0.3, and g = 0.2 according to actual test results. For example, for the candidate sound effect "human voice enhancement + moderate bass enhancement", the user preference score is 0.75, the scene adaptation score is 0.8, and the sound effect popularity score is 0.7, and the comprehensive score is 0.5 x 0.75 + 0.3 x 0.8 + 0.2 x 0.7 = 0.755.

[0065] In the above embodiment, the recommendation algorithm is pre-set, and the candidate sound effect set can be sorted according to the pre-set recommendation algorithm, the optimal sound effect is recommended to the user, and the recommendation algorithm can be optimized according to the user feedback. The above-mentioned pre-set recommendation algorithm considers the user's preference degree for different sound effects, the matching degree of the current scene and the sound effect, and the popularity of the sound effect in the user group. A corresponding weight is set for each factor, for example, the user preference degree weight is 0.5, the scene matching degree weight is 0.3, and the popularity weight is 0.2. For each sound effect in the candidate sound effect set, the comprehensive score is calculated according to the weight value of the factor, and the sound effects are sorted in descending order of the score. The sound effect with the highest score is selected as the recommended sound effect and is highlighted on the interface of the sound effect device for the user.

[0066] In the above embodiment, the optimization of the recommendation algorithm according to the user feedback includes: taking the user feedback, the current recommended scene, and the user preference data as new training data, re-inputting them into the optimization model of the recommendation algorithm for training, and adjusting the parameters of the recommendation algorithm model.

[0067] For example, the user is not satisfied with the recommended sound effect and clicks the "dislike" feedback button. The system records the user feedback information, takes the user feedback, the current recommended scene, and the user preference data as new training data, re-inputs them into the optimization model of the recommendation algorithm for training, adjusts the parameters of the recommendation algorithm, and improves the accuracy of subsequent recommendations.

[0068] Further, for the optimization of the recommendation algorithm, user feedback data can be periodically (weekly) analyzed, and a reinforcement learning algorithm (such as the Q-learning algorithm) can be used to adjust the weight coefficients (a, b, g) in the recommendation algorithm and the scene-sound effect association rule base. For example, if it is found that the user's satisfaction score for sound effects with high scene adaptation is generally high, the value of b can be appropriately increased; if the user feedback indicates that a certain sound effect mode is not matched with the scene in actual use, the association rule can be deleted from the scene-sound effect association rule base. By continuously optimizing the recommendation algorithm, the accuracy of subsequent sound effect recommendations and user satisfaction can be improved.

[0069] Embodiment 1: A sound effect recommendation method based on user preferences and scene perception is proposed according to the specific embodiments described above, which provides users with a more realistic audio experience that fits the actual use scenario. It can be summarized in the following aspects:

[0070] 1. User preference data information collection

[0071] Through the user's historical operation records on the vehicle audio equipment or mobile phone application, including but not limited to music file information, play frequency, stay time, manually selected sound effect mode, etc., the user's basic preference data is collected. At the same time, a user-defined preference setting interface is provided, allowing users to manually input their preferences for different music styles, instruments, vocals, etc., further enriching the user preference data. As can be seen, by deeply analyzing the user's historical operation records and manual preference settings, a precise user preference model can be constructed, which can fully consider the user's individual needs and recommend sound effects that meet the user's unique taste, improving user satisfaction with sound effect recommendations.

[0072] 2. Scene perception

[0073] The vehicle entertainment system host and vehicle sensors, such as acceleration sensors, gyroscopes, sunlight and rainfall sensors, microphones, etc., are used to obtain vehicle location information, navigation information, driving state, ambient light intensity, ambient noise level, etc., to determine the current scenario. For example, the vehicle position and driving state are determined based on the data from the acceleration sensor and gyroscope; the environment is determined to be rainy or sunny based on the sunlight and rainfall sensor; and the ambient noise in the vehicle is distinguished based on the ambient noise intensity collected by the microphone, with the assistance of cloud data from the vehicle.

[0074] 3. Preference analysis and scene matching

[0075] The collected user preference data is deeply analyzed to construct a user preference model. For example, the preference weights of the user for different music styles and sound effect types are analyzed by a machine learning algorithm. At the same time, a scene and sound effect association model is established to determine the appropriate sound effect type in different scenes. Then, the current perceived scene is matched with the user preference model to find a candidate sound effect set that is most suitable for the current scene and user preference.

[0076] 4. Sound effect recommendation

[0077] According to a preset recommendation algorithm, the sound effects in the candidate sound effect set are sorted, the sound effects at the top of the sorting are selected as recommended sound effects, and the recommended sound effects are displayed to the user. The recommendation algorithm can comprehensively consider the user's preference degree for different sound effects, the matching degree of the current scene and the sound effect, the popularity of the sound effect, and other factors. If the user is not satisfied with the recommended sound effect, the system records the user feedback to further optimize the recommendation algorithm and improve the accuracy of subsequent recommendations.

[0078] The above technical content can automatically recommend sound effects for the user, without the user manually selecting from a large number of sound effect options, greatly simplifying the user's operation process of selecting sound effects, saving the user's time, and improving the user's efficiency of using the audio device. On the other hand, the recommendation algorithm can be continuously optimized according to the user's feedback on the recommended sound effect, so that the recommendation result is more accurate, and the recommendation effect is better and better with the increase of the user's use times, providing the user with better and better audio experience. Based on the same inventive concept, the embodiments of the present application also provide a sound effect recommendation system based on user preference and scene perception. The implementation scheme for solving the problem provided by the system is similar to the implementation scheme described in the above method, so the specific limitations in one or more sound effect recommendation system embodiments based on user preference and scene perception provided below can be referred to the limitations of the above sound effect recommendation method based on user preference and scene perception, which will not be repeated here.

[0079] In one embodiment, a sound effect recommendation system based on user preference and scene perception is provided, as shown in Figure 2 which includes a model construction module 210, a determination module 220, a matching module 230, a screening module 240, and a demand evaluation unit 250, wherein:

[0080] The model construction module 210 is configured to deeply analyze the collected user preference data and construct a user preference model.

[0081] The determination module 220 is configured to, when receiving an audio recommendation request of a user terminal, acquire scene perception information matching a current vehicle position, and determine a current scene.

[0082] The matching module 230 is configured to match the current scene with the user preference model, and obtain a candidate sound effect set matched with the current scene and the user preference;

[0083] The screening module 240 is configured to view each candidate sound effect data in the candidate sound effect set, screen the candidate sound effect data according to the sound effect data, and generate a recommendation result.

[0084] The recommendation module 250 is configured to feed back the sound effect recommendation result.

[0085] Based on the same inventive concept, the present application also provides a computer device, comprising at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method according to any one of claims S101-S105.

[0086] A computer readable storage medium, which stores computer program instructions, when the computer program instructions are executed by a processor, implement the method according to any one of claims S101-S105.

[0087] In one embodiment, the computer device provided by the present application can be a terminal, and its internal structure diagram can be as shown in Figure 3 The computer device comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved through WIFI, mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement a sound effect recommendation method based on user preference and scene perception.

[0088] Those skilled in the art can understand that Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can comprise more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0089] It should be noted that any technical features in the above embodiments can be combined in any manner, and for the sake of brevity, not all possible combinations are described, but it is understood that any combination of the technical features is within the scope of the present disclosure.

[0090] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be noted that for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these are within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A sound effect recommendation method based on user preference and scene perception, characterized in that: The method comprises: Conduct in-depth analysis of collected user preference data and build a user preference model; When receiving an audio recommendation request from the user, obtain scene perception information matching the current vehicle position and determine the current scene; Match the current scene with the user preference model to obtain a set of candidate sound effects that match the current scene and user preferences; Checking each candidate sound effect data in the candidate sound effect set, screening the candidate sound effect data according to the sound effect data, and generating a recommendation result; Feedback the sound effect recommendation result including the recommended sound effects.

2. The method according to claim 1, wherein The collecting of user preference data includes: collecting historical operation data of the user on the vehicle audio device and preference setting information manually input by the user; User preference data is obtained based on the historical operation records of the user on the vehicle audio device and the preference setting information manually input by the user.

3. The method according to claim 2, wherein The historical operation records of the matching user information include: historical operation records of the user on the vehicle audio device or mobile terminal application, including the played music file information, play frequency, stay time, and manually selected sound effect mode; The user's behavioral data includes: the user's preferences for different music styles, instruments, and vocals input through the vehicle-mounted interactive interface.

4. The method according to claim 1, wherein The in-depth analysis of the collected user preference data and the construction of the user preference model include: using a machine learning algorithm to learn the correlation between the user's historical operation records on the vehicle audio equipment and the preference setting information manually input by the user, and analyzing the user's preference for different music styles and sound effect types; comprehensively considering the user's preference for different sound effects, the matching degree between the current scene and the sound effect, and the popularity of the sound effect among the user group; and setting corresponding weights for each factor to establish a user preference model.

5. The method according to claim 1, wherein The acquiring of scene perception information matching the current vehicle position includes: Locate the current position of the user's vehicle and display all scene perception information covering the current position of the user's vehicle within the preset area, including vehicle position information, navigation information, driving status, ambient light intensity, and ambient noise level information; The scene perception information covering the current position of the user's vehicle is sorted in sequence according to predefined rules, and feature comparison is performed to identify the scene the vehicle is currently in.

6. The method according to claim 1, wherein The method further includes: calculating a comprehensive score for each sound effect in the candidate sound effect set based on the factor weight, and sequentially sorting the sound effects from high to low according to the score to obtain a sound effect recommendation result; the sound effect recommendation result includes at least one optimal sound effect; The optimal sound effect is recommended to the user, and the recommendation algorithm is optimized based on user feedback.

7. The method according to claim 6, wherein Optimizing the recommendation algorithm based on user feedback includes: using user feedback, current recommendation scenarios, user preference data, etc. as new training data, re-inputting them into the optimization model of the recommendation algorithm for training, and adjusting the parameters of the recommendation algorithm model.

8. A sound effect recommendation system based on user preference and scene perception, characterized in that: The system comprises: Model building module, used to conduct in-depth analysis of collected user preference data and build a user preference model; A determination module, configured to, upon receiving an audio recommendation request from a user terminal, obtain scene perception information matching the current vehicle position and determine the current scene; A matching module is used to match the current scene with the user preference model and obtain a set of candidate sound effects that match the current scene and user preferences; a screening module, configured to view each candidate sound effect data in the candidate sound effect set, screen the candidate sound effect data based on the sound effect data, and generate a recommendation result; The recommendation module is used to feed back the recommended sound effects included in the sound effect recommendation result.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of a sound effect recommendation method based on user preference and scene perception as described in any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of a sound effect recommendation method based on user preference and scene perception as described in any one of claims 1 to 7.

Images