Vehicle-mounted audio control method, system, equipment and medium

Abstract

The invention discloses a vehicle-mounted audio control method, system and device and a medium, and the method comprises the steps: matching the user information of a passenger at each current seat through the weight of the passenger at each current seat, matching a first song list for the passenger through a pre-training model, and if the information of the passenger at the current seat cannot be matched, carrying out the matching through a physiology-behavior combined clustering method, the current seat passenger person information and the cloud user clustering center are matched to the most similar group, a song list with the highest similarity between the group and the current seat passenger person information is extracted as a second song list, and during song list playing, a pressure vibration sensor is monitored to obtain the body movement frequency of each current seat passenger; analyzing the body movement frequency of the passengers on each current seat so as to obtain user feedback information, obtaining respective preference list overlapping parts of the passengers on each current seat by using the user feedback information, and generating a third song list for playing; the problem that personalized playing control of an existing vehicle-mounted audio playing device in a multi-passenger scene is insufficient is solved.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology and relates to a vehicle-mounted audio control method, system, device and medium. Background Art

[0002] With the continuous development of the automotive industry and people's increasing demands for a better driving experience, in-car audio equipment has become an indispensable part of modern vehicles. It not only provides drivers and passengers with rich entertainment content to alleviate the boredom of driving, but also enhances driving safety and comfort by playing audio information such as music and news.

[0003] Currently, most in-car audio devices on the market control music playback based on the driver's or the car's associated music preferences. This approach can better meet the user's personalized needs when the driver is driving alone or the vehicle is primarily used by a specific person. However, in real-world scenarios, a vehicle, as a means of transportation, often carries more than one person. This is especially true for family trips, gatherings with friends, or carpooling trips, where multiple passengers may have significantly different musical preferences.

[0004] Existing in-car audio control systems struggle to effectively collect and analyze the music preferences of multiple passengers. Lacking a precise understanding of shared preferences, the system often recommends tracks based on the preferences of the driver or a single linked account, potentially failing to satisfy the tastes of most passengers. This not only degrades the passengers' audio entertainment experience but also significantly reduces the level of personalization of in-car audio devices.

[0005] Therefore, how to improve the personalized playback control capabilities of in-vehicle audio playback equipment in multi-passenger scenarios so that it can recommend playback tracks based on the common preferences of multiple passengers has become an important issue that needs to be urgently addressed in the current field of in-vehicle audio playback technology. Summary of the Invention

[0006] To solve the problems existing in the background technology, the present invention proposes a vehicle-mounted audio control method, system and device, which aims to solve the problem that existing vehicle-mounted audio playback equipment lacks personalized playback control in multi-passenger scenarios.

[0007] A first aspect of the present application provides a vehicle audio control method, comprising:

[0008] Use the weight of the current passengers in each seat to match the user information of the current passengers in each seat, and use the pre-trained model to match the first playlist for the passengers;

[0009] If there is no matching information for the current seat passenger, the current seat passenger's character information is obtained through a physiological behavior joint clustering method, and matched with the cloud user cluster center to the most similar group, and the playlist with the highest similarity between the group and the current seat passenger's character information is extracted as the second playlist;

[0010] During the playback of the first playlist or the second playlist, the pressure vibration sensor is monitored to obtain the body movement frequency of the passengers in each seat, and the body movement frequency of the passengers in each seat is analyzed to obtain user feedback information;

[0011] The user feedback information is used to obtain the preference lists of the passengers in each seat at the current time, and a third song list is generated for playback based on the overlapping parts of the preference lists.

[0012] Optionally, construct the current seat passenger's character information, including:

[0013]

[0014] μ v is the average vehicle speed, is the vehicle speed variance, μ Δv is the average value of vehicle speed change information, is the variance of vehicle speed change information, μt is the average frequency of passenger body movement, is the passenger's body motion frequency variance, μ Δt is the average value of the change information of the passenger's body movement frequency, is the variance of the passenger's body movement frequency change information, is the main frequency of the vehicle's speed, The main frequency is the passenger's body movement.

[0015] Optionally, match to the most similar group, including:

[0016]

[0017] F new is the character model of the current seat passenger, F i The i-th person model in the cloud data, d is the total number of features, w j is the weight of the j-th feature, F new，j is the jth feature of the character model of the current seat passenger, F i，j is the jth feature of the i-th person model in the cloud data.

[0018] Optionally, the process of generating the second playlist includes:

[0019] Reading the playlists corresponding to the most similar groups to obtain the intersection of the playlists corresponding to the most similar groups;

[0020] A second playlist is generated according to the intersection of the corresponding playlists in the most similar groups.

[0021] Optionally, the monitoring of the pressure vibration sensor to obtain the body movement frequency of the current passengers in each seat includes:

[0022] The synchronization rate R between the user's movement frequency F and the song rhythm B is calculated through motion rhythm coupling analysis:

[0023]

[0024] in, is the phase extraction function, R>0.7 is marked as liked, R≤0.7 is disliked, T is the total playing time of the first playlist or the second playlist, t is the time period for any song to be played, F(t) is the user's movement frequency in the t time period, and B(t) is the rhythm of the song in the t time period.

[0025] Optionally, the generation process of the third playlist is:

[0026] Using a reinforcement learning strategy, the song weights are updated according to the synchronization rate R to generate the third playlist:

[0027]

[0028] in, For the third song list, is the first or second playlist, α is the learning rate, R is the synchronization rate, β is the search factor, Entropy is the entropy function, Q song Any song in the first or second playlist.

[0029] Optionally, after generating the third playlist, the following steps are further included:

[0030] The pre-trained model adjusts the third playlist based on the feedback information of the passengers in each seat on the third playlist, generates multiple target playlists, and generates a fourth playlist based on the intersection of the multiple target playlists.

[0031] In a second aspect of the present application, a vehicle-mounted audio control system is provided, comprising:

[0032] The first module is used to match the user information of the current passengers in each seat with the weight of the current passengers in each seat, and use the pre-trained model to match the first playlist for the passengers;

[0033] The second module is configured to obtain the current seat passenger's character information through a physiological behavior joint clustering method if there is no matching current seat passenger information, match it with the cloud user cluster center to the most similar group, and extract the playlist with the highest similarity between the group and the current seat passenger's character information as the second playlist;

[0034] The third module is configured to monitor the pressure vibration sensor to obtain the body movement frequency of passengers in each seat during the playback of the first and second playlists, and analyze the body movement frequency of passengers in each seat to obtain user feedback information;

[0035] The fourth module is used to use the user feedback information to obtain the preference lists of the passengers in each seat at present, and to generate a third playlist for playback based on the overlapping parts of the preference lists.

[0036] In a third aspect of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned in-vehicle audio control method when executing the computer program.

[0037] In a fourth aspect of the present application, a computer-readable medium is provided, wherein the computer-readable medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned in-vehicle audio control method is implemented.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The aforementioned in-vehicle audio control method, system, and device first utilize the weight of each seat's current occupant to match user information, combined with a pre-trained model to accurately match a first playlist for each passenger. This fully considers the passenger's initial characteristics, ensuring a high degree of personalized music recommendations from the initial stage. Then, for passenger information that cannot be directly matched, a physiological and behavioral joint clustering method is used to match the passenger's personal information with cloud-based user cluster centers, and the playlist with the highest similarity is extracted as the second playlist. This further broadens the coverage of personalized recommendations, ensuring that even with incomplete information, passengers can still be provided with relatively appropriate music choices, significantly improving the accuracy of personalized music recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a flow chart of a vehicle audio control method according to an embodiment of the present invention;

[0041] Figure 2 is a schematic diagram of a vehicle audio control system according to an embodiment of the present invention;

[0042] Figure 3Schematic diagram of a robot production safety supervision device in one embodiment of the present invention. DETAILED DESCRIPTION

[0043] 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.

[0044] In one embodiment, if Figure 1 As shown, a vehicle audio control method is provided, which is applied in Figure 1 Take the example of , and explain it, including the following specific steps:

[0045] S10: Use the weight of the current passengers in each seat to match the user information of the current passengers in each seat, and match the first playlist for the passengers through the pre-trained model.

[0046] Specifically, every seat in a vehicle may have a passenger. Pressure and vibration sensors installed on the seats capture the weight of each passenger. This weight data is then matched against pre-stored passenger user information. This passenger user information includes the passenger's name, age, gender, previous ride history, and personal preferences. Using weight as a reference factor, the corresponding passenger's specific information is found. For example, if a certain weight range corresponds to a specific passenger, the specific passenger in that seat can be determined.

[0047] The pre-trained model is a machine learning or deep learning model trained with a large amount of data. This pre-trained model is combined with matched passenger user information, such as age, gender, and personal music preferences, to recommend a suitable playlist for each passenger. This playlist, known as the first playlist, is composed of a series of songs predicted by the pre-trained model to be the passenger's most likely favorite. This playlist is played during the ride, enhancing the passenger's experience.

[0048] For example, when a passenger gets in a vehicle and sits down, the seat's pressure and vibration sensors detect changes in pressure in real time, thereby calculating the passenger's weight. For example, if the passenger in the front passenger seat is detected to weigh 60 kg, the vehicle's intelligent system will search its user database, which contains information about passenger Xiao Ming (weight range 5862 kg). Based on weight matching, the system identifies the passenger currently sitting in the front passenger seat as Xiao Ming. Xiao Ming's user information also includes age 30, male, a preference for pop and rock music, and a history of frequently listening to Jay Chou and Mayday while riding in the vehicle. The vehicle system has a pre-trained music recommendation model based on deep learning. This model selects a series of songs from a vast music library based on Xiao Ming's age, gender, and music preferences to create a playlist. For example, the model will select songs such as "Blue and White Porcelain" and "Qilixiang" by Jay Chou, and "Zhizu" and "Jiangqiang" by Mayday, generating a 20-song playlist as Xiao Ming's first match. Then, the vehicle's audio system will start playing the songs in this playlist, allowing Xiao Ming to enjoy his favorite music while riding.

[0049] The same process will also be applied to passengers in other seats in the car. Each passenger’s user information will be matched according to their weight, and then a personalized first playlist will be matched for them through the pre-trained model to improve the overall riding experience.

[0050] S20: If there is unmatched current seat passenger information, the current seat passenger character information is obtained through a physiological behavior joint clustering method, matched with the cloud user clustering center to the most similar group, and the playlist with the highest similarity between the group and the current seat passenger character information is extracted as the second playlist.

[0051] Specifically, during vehicle operation, after obtaining passenger weight and other information through seat-mounted pressure and vibration sensors, the existing local user database cannot find matching passenger information. For example, if a new passenger is riding the vehicle for the first time, there may be no relevant record in the system, or if there are errors in weight measurement, the existing user information cannot be accurately matched. The physiological and behavioral joint clustering method then comprehensively considers the passenger's physiological and behavioral information and classifies and aggregates the passenger information. Because cluster information for a large amount of users is stored in the cloud, each cluster center represents a group of users with similar characteristics. The physiological and behavioral joint information of the currently unmatched passenger is compared with the cluster centers in the cloud to identify the user group with the most similar characteristics. The corresponding playlists of the most similar group are then retrieved to obtain the intersection of the corresponding playlists in the most similar group. A second playlist is generated based on the intersection of the corresponding playlists in the most similar group. The playlist corresponding to the most similar group is a group composed of multiple specific playlists, and the common components of the group composed of specific playlists are obtained to form the second playlist. For example, if a passenger's physiological and behavioral characteristics are most similar to the "young people" group in the cloud, the passenger will be classified into this group.

[0052] In an embodiment of the present application, the passenger's liking for the song is judged based on the passenger's physiological characteristics, such as weight, and behavioral characteristics such as the state of body movement while the song is playing and the car driving status, and similar songs are searched to form a second playlist.

[0053] The frequency of a passenger's body movements includes the frequency of body movements and the amplitude of changes in body movement frequency. By calculating the average body movement frequency, the variance of the body movement frequency, the average body movement frequency variation information, and the variance of the body movement frequency variation information, we can comprehensively obtain the passenger's accurate body movement information.

[0054] For example, by calculating the average of a passenger's body movement frequency, we can determine their average level of restlessness and activity. A higher average of a passenger's body movement frequency indicates a preference for lively and upbeat songs, while a lower average indicates a preference for quieter and more soothing songs. By calculating the variance of a passenger's body movement frequency, we can determine how quickly their mood changes. A higher variance indicates a faster mood swing and a greater adaptability to songs of varying styles. A lower variance indicates a slower mood swing and a lower adaptability to songs of varying styles. The average of a passenger's body movement frequency variation can also be used to determine the magnitude of their mood swings. A higher average of their body movement frequency variation indicates a greater mood swing and a greater tolerance for a variety of song styles. A lower average of their body movement frequency variation indicates a lower mood swing and a narrower range of song styles. The variance of the passenger's body movement frequency is used to reflect the passenger's emotional fluctuation frequency. If the variance of the passenger's body movement frequency is large, it indicates that the passenger's emotional fluctuation frequency is large, which means that the frequency of changes in the different song styles in the passenger's playlist is large. If the variance of the passenger's body movement frequency is small, it indicates that the passenger's emotional fluctuation frequency is small, which means that the frequency of changes in the different song styles in the passenger's playlist is small. In this application, the magnitude of the body movement frequency is related to the passenger's weight collected by the passenger seat pressure vibration sensor, but it does not affect the high-precision collection of the magnitude of the body movement frequency by existing technical means.

[0055] Passenger driving behavior includes information about vehicle speed fluctuations, acceleration frequency, and acceleration amplitude. For example, by calculating the average vehicle speed, we can determine whether a passenger is impatient or patient during driving. A high average speed indicates that the passenger prefers fast-paced songs, while a low average speed indicates that the passenger prefers slow-paced songs. By calculating the variance of vehicle speed, we can determine the degree of variability during driving, which in turn indicates the passenger's acceptance of change. A high variance indicates a high level of acceptance and a wide range of song styles. A low variance indicates a low level of acceptance and a limited or single-style acceptance. By calculating the average speed variation information, we can also determine the passenger's desire for excitement. A high variance in vehicle speed indicates a high preference for exciting and thrilling songs, while a low variance indicates a high preference for steady, slow-paced songs. By calculating the variance of vehicle speed change information, we can determine the passenger's acceptance of stimuli. If the variance of vehicle speed change information is large, it means that the passenger has a strong acceptance of stimuli, and different types of stimulating songs can be recommended. If the variance of vehicle speed change information is small, it means that the passenger has a weak acceptance of stimuli, and fewer or a single type of songs can be recommended.

[0056] It is worth noting that in an embodiment of the present application, the passenger's body movement frequency average value, the passenger's body movement frequency change information variance, the passenger's body movement frequency variance, and the passenger's body movement frequency change information variance are used to comprehensively judge the passenger's physiological characteristics in terms of his or her liking for the songs in the first or second playlist being played. In combination with driving behavior information, such as calculating the vehicle's driving speed average value, the vehicle's driving speed variance, the vehicle's driving speed change information average value, and the vehicle's driving speed change information variance, the driver's and passenger's behavioral characteristics are obtained. The passenger's liking for the playlist being played is comprehensively judged through physiological and behavioral characteristics, and the liking level is fed back to the vehicle system to adjust the playlist.

[0057] After finding the most similar group, that group may already have multiple different playlists. The intersection of the playlists corresponding to the most similar group is selected from these playlists and used as the second playlist recommended to the passenger in the current seat. This playlist is played during the ride to meet the passenger's music needs.

[0058] The specific method of obtaining the current seat passenger information is to calculate using the following formula:

[0059]

[0060] Among them, μv is the average vehicle speed, is the vehicle speed variance, μ Δv is the average value of vehicle speed change information, is the variance of vehicle speed change information, μ t is the average frequency of passenger body movements, is the passenger's body motion frequency variance, μ Δt is the average value of the change information of the passenger's body movement frequency, is the variance of the passenger's body movement frequency change information, The main frequency of the vehicle's speed is obtained by Fourier transforming the frequency domain characteristics of the vehicle's speed, converting the vehicle's speed from the time domain information to the frequency domain, and providing data support for the average vehicle speed, the variance of the vehicle's speed, the average vehicle speed change information, and the variance of the vehicle's speed change information, to avoid the lack of perception of changes in passenger data when the amount of data is large. The main frequency of the passenger's body movement is obtained by Fourier transform, and the passenger's body movement information is converted from the time domain to the frequency domain. The average value of the passenger's body movement frequency, the variance of the passenger's body movement frequency, the average value of the passenger's body movement frequency change information and the variance of the passenger's body movement frequency change information are supported by data to avoid the lack of perception of changes in passenger data when the amount of data is large.

[0061] The specific method of matching to the most similar group is calculated by the following formula:

[0062]

[0063] F new is the character model of the current seat passenger, F i The i-th person model in the cloud data, d is the total number of features, w j is the weight of the j-th feature, F new，j is the jth feature of the character model of the current seat passenger, F i，j is the jth feature of the i-th person model in the cloud data.

[0064] S30: During the playback of the first playlist or the second playlist, the pressure vibration sensor is monitored to obtain the body movement frequency of the passengers in each seat, and the body movement frequency of the passengers in each seat is analyzed to obtain user feedback information.

[0065] Specifically, when the vehicle's audio system plays a first playlist generated by a pre-trained model after matching known passenger information, or a second playlist generated by a joint physiological behavior clustering method for unmatched passengers, pressure vibration sensors installed on the vehicle's seats, capable of detecting body movement, can monitor the passenger's body movements in real time. Body movement frequency refers to the number or rhythm of a passenger's movements over a certain period of time. For example, if a passenger gently sways their body to the rhythm of the music, the pressure vibration sensor can capture this movement and record the corresponding frequency data. The body movement frequency data collected by the pressure vibration sensor for each seat is then analyzed in depth. This data can be used to infer feedback, such as passenger preference and satisfaction with the playlist being played. For example, if a passenger's body movement frequency matches the rhythm of the music and is frequent, it may indicate that the passenger enjoys the playlist; conversely, if the passenger barely moves, it may indicate that the playlist is not very interesting.

[0066] For example, when a popular song with a strong rhythm is played, the pressure vibration sensor detects that the driver's body sways slightly to the rhythm of the music. Calculations show that the driver's body movement frequency roughly matches the rhythm of the music. Meanwhile, the friend sitting in the passenger seat has almost no noticeable body movement, with a body movement frequency close to 0 beats per minute. The vehicle's intelligent system analyzes this body movement frequency data. Based on the analysis results, it determines that the driver is relatively satisfied with the current playlist because their body movement frequency echoes the rhythm of the music, indicating a positive response. Meanwhile, the friend may not be very interested in the playlist because they have hardly any body movement. Based on this feedback, the vehicle system can further adjust the playlist to the types of music that both the driver and the passenger seat enjoy, thereby enhancing the overall riding experience and musical enjoyment.

[0067] In this application, the synchronization rate R between the user's movement frequency F and the song rhythm B is calculated through motion rhythm coupling analysis:

[0068]

[0069] in, is the phase extraction function, R>0.7 is marked as liked, R≤0.7 is disliked, T is the total playing time of the first playlist or the second playlist, t is the time period for any song to be played, F(t) is the user's movement frequency in the t time period, and B(t) is the rhythm of the song in the t time period.

[0070] S40: Utilize the user feedback information to obtain the preference lists of the passengers in each seat, and generate a third playlist for playback based on the overlapping parts of the preference lists.

[0071] Specifically, based on the user feedback from passengers in each seat when a music playlist is played, a music preference list can be generated for each seat. This list contains information about the passenger's likely favorite music genres, artists, and songs. The preference lists of all passengers in each seat are then compared to identify any overlaps, i.e., music-related elements that all passengers share. For example, multiple passengers may have the "pop" genre or the names of certain artists on their preference lists. Based on the overlaps found in the preference lists of each seat, corresponding songs are selected from the vehicle's music library to form a new playlist, the third playlist. This third playlist is then played through the vehicle's audio system to satisfy the shared musical preferences of all passengers and enhance the overall in-vehicle music experience.

[0072] It is worth noting that this application uses a reinforcement learning strategy to update the song weights according to the synchronization rate R to generate the third playlist:

[0073]

[0074] in, For the third song list, is the second song list, α is the learning rate, R is the synchronization rate, β is the search factor, Entropy is the entropy function, Q song Any song in the second playlist.

[0075] In one embodiment, in step S40, i.e., after generating the third playlist, the following steps are further included:

[0076] The pre-trained model adjusts the third playlist based on the feedback information of the passengers in each seat on the third playlist, generates multiple target playlists, and generates a fourth playlist based on the intersection of the multiple target playlists.

[0077] Specifically, while the third playlist is playing, pressure and vibration sensors are used to obtain information such as the body movement frequency of passengers in each seat, which serves as passenger feedback on the third playlist. The pre-trained model then uses this feedback to understand passengers' preferences and satisfaction with the songs on the third playlist. Based on this feedback, the pre-trained model modifies and adjusts the third playlist. For example, if certain songs are found to be disliked by passengers, these songs are removed from the playlist and new songs that may suit the passengers' preferences are added. Using different adjustment methods and strategies, multiple target playlists are generated. Each target playlist may differ in song selection and order, but all are optimized based on passenger feedback on the third playlist. The generated target playlists are then compared to identify songs that are common across all of them. This common set of songs is the intersection of the target playlists. Based on this intersection, a new playlist, the fourth playlist, is generated. This fourth playlist includes songs from all target playlists and, in theory, better meets the common preferences of passengers in each seat.

[0078] For example, the vehicle system plays a third playlist for the driver and five passengers. During playback, seat-mounted pressure and vibration sensors and other possible monitoring devices are used to obtain feedback from each passenger on the third playlist. For example, the driver is not very interested in one of the slower songs and barely moves to the music; Passenger 1 is very fond of a pop song on the playlist, humming along and swaying his body frequently; Passenger 2 reacts strongly to a rock song on the playlist, expressing excitement. The pre-trained model adjusts the third playlist based on this feedback. The slow song that the driver dislikes is removed from the playlist and a classic song with a moderate tempo is added. For the pop song that Passenger 1 likes, several other pop songs of the same genre and by the same artist are added; for the rock song that Passenger 2 likes, several rock songs of a similar style are added. Through different adjustments, three different target playlists are generated. Target playlist 1 is tailored to the driver's preferences, while Target playlist 2 focuses on satisfying the pop music preferences of Passenger 1, and Target playlist 3 caters to rock music enthusiasts like Passenger 2. Comparing these three target playlists revealed that they all shared a classic pop song, "Friends," and an upbeat rock track, "Shame." Based on these two shared songs, a fourth playlist of eight songs was created by selecting songs from the music library with similar styles or wider audiences. The vehicle's audio system then began playing this fourth playlist. Because it incorporates the common elements of multiple target playlists, it is more likely to satisfy the musical preferences of all passengers, enhancing their overall ride experience.

[0079] In one embodiment, if Figure 2 As shown, a vehicle audio control system is provided. The vehicle audio control system corresponds one-to-one to the vehicle audio control method in the above embodiment. The vehicle audio control system includes: a first module, a second module, a third module, and a fourth module. The functional modules are described in detail as follows:

[0080] The first module is used to match the user information of the current passengers in each seat with the weight of the current passengers in each seat, and use the pre-trained model to match the first playlist for the passengers;

[0081] The second module is configured to obtain the current seat passenger's character information through a physiological behavior joint clustering method if there is no matching current seat passenger information, match it with the cloud user cluster center to the most similar group, and extract the playlist with the highest similarity between the group and the current seat passenger's character information as the second playlist;

[0082] The third module is configured to monitor the pressure vibration sensor to obtain the body movement frequency of passengers in each seat during the playback of the first and second playlists, and analyze the body movement frequency of passengers in each seat to obtain user feedback information;

[0083] The fourth module is used to use the user feedback information to obtain the preference lists of the passengers in each seat at present, and to generate a third playlist for playback based on the overlapping parts of the preference lists.

[0084] The specific definitions of the in-vehicle audio control system can be found in the definitions of the in-vehicle audio control method above and will not be further elaborated here. Each module in the aforementioned in-vehicle audio control system may be implemented in whole or in part through software, hardware, or a combination thereof. Each of these modules may be embedded in or independent of a processor within a computer device in hardware form, or may be stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0085] In one embodiment, Figure 3 As shown, an electronic 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, a vehicle audio control method is implemented.

[0086] For the specific limitations on electronic devices, please refer to the limitations on the in-vehicle audio control method above, which will not be repeated here.

[0087] In one embodiment, a computer-readable medium is provided, wherein the computer-readable medium stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0088] Use the weight of the current passengers in each seat to match the user information of the current passengers in each seat, and use the pre-trained model to match the first playlist for the passengers;

[0089] If there is no matching information for the current seat passenger, the current seat passenger's character information is obtained through a physiological behavior joint clustering method, and matched with the cloud user cluster center to the most similar group, and the playlist with the highest similarity between the group and the current seat passenger's character information is extracted as the second playlist;

[0090] During the playback of the first playlist or the second playlist, the pressure vibration sensor is monitored to obtain the body movement frequency of the passengers in each seat, and the body movement frequency of the passengers in each seat is analyzed to obtain user feedback information;

[0091] The user feedback information is used to obtain the preference lists of the passengers in each seat at the current time, and a third song list is generated for playback based on the overlapping parts of the preference lists.

[0092] For specific definitions of computer-readable media, please refer to the above definitions of the in-vehicle audio control method, which will not be repeated here.

[0093] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods 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 can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an 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 (Synchl ink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM). Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0094] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A vehicle audio control method, characterized in that: Includes: Use the weight of the current passengers in each seat to match the user information of the current passengers in each seat, and use the pre-trained model to match the first playlist for the passengers; If there is no matching information for the current seat passenger, the current seat passenger's character information is obtained through a physiological-behavioral joint clustering method, and matched with the cloud user cluster center to the most similar group, and the playlist with the highest similarity between the group and the current seat passenger's character information is extracted as the second playlist; During the playback of the first playlist or the second playlist, the pressure vibration sensor is monitored to obtain the body movement frequency of the passengers in each seat, and the body movement frequency of the passengers in each seat is analyzed to obtain user feedback information; The user feedback information is used to obtain the preference lists of the passengers in each seat at the current time, and a third song list is generated for playback based on the overlapping parts of the preference lists.

2. The vehicle audio control method according to claim 1, characterized in that: Get the current seat passenger information, including: μ v is the average vehicle speed, is the vehicle speed variance, μ Δv is the average value of vehicle speed change information, is the variance of vehicle speed change information, μ t is the average frequency of passenger body movements, is the passenger's body motion frequency variance, μ Δt is the average value of the change information of the passenger's body movement frequency, is the variance of the passenger's body movement frequency change information, is the main frequency of the vehicle's speed, The main frequency is the passenger's body movement.

3. The vehicle audio control method according to claim 1, characterized in that: Match to the most similar groups, including: F new is the character model of the current seat passenger, F i The i-th person model in the cloud data, d is the total number of features, w j is the weight of the j-th feature, F new，j is the jth feature of the character model of the current seat passenger, F i，j is the jth feature of the i-th person model in the cloud data.

4. The vehicle audio control method according to claim 1, characterized in that: The process of generating the second playlist includes: Reading the playlists corresponding to the most similar groups to obtain the intersection of the playlists corresponding to the most similar groups; A second playlist is generated according to the intersection of the corresponding playlists in the most similar groups.

5. The vehicle audio control method according to claim 1, characterized in that: The monitoring pressure vibration sensor obtains the body movement frequency of the current passengers in each seat, including: The synchronization rate R between the user's movement frequency F and the song rhythm B is calculated through motion-rhythm coupling analysis: in, is the phase extraction function, R>0.7 is marked as liked, R≤0.7 is disliked, T is the total playing time of the first playlist or the second playlist, t is the time period for any song to be played, F(t) is the user's movement frequency in the t time period, and B(t) is the rhythm of the song in the t time period.

6. The vehicle audio control method according to claim 1, characterized in that: The generation process of the third song list is as follows: Using a reinforcement learning strategy, the song weights are updated according to the synchronization rate R to generate the third playlist: in, For the third song list, is the first or second playlist, α is the learning rate, R is the synchronization rate, β is the search factor, Entropy is the entropy function, Q song Any song in the first or second playlist.

7. The vehicle audio control method according to claim 1, characterized in that: After generating the third playlist, it also includes: The pre-trained model adjusts the third playlist based on the feedback information of the passengers in each seat on the third playlist, generates multiple target playlists, and generates a fourth playlist based on the intersection of the multiple target playlists.

8. A vehicle audio control system, characterized in that: Includes the following modules: The first module is used to match the user information of the current passengers in each seat with the weight of the current passengers in each seat, and use the pre-trained model to match the first playlist for the passengers; The second module is configured to obtain the current seat passenger's character information through a physiological-behavioral joint clustering method if there is no matching current seat passenger information, match it with the cloud user cluster center to the most similar group, and extract the playlist with the highest similarity between the group and the current seat passenger's character information as the second playlist; The third module is configured to monitor the pressure vibration sensor to obtain the body movement frequency of passengers in each seat during the playback of the first and second playlists, and analyze the body movement frequency of passengers in each seat to obtain user feedback information; The fourth module is used to use the user feedback information to obtain the preference lists of the passengers in each seat at present, and to generate a third playlist for playback based on the overlapping parts of the preference lists.

9. An electronic 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 in-vehicle audio control method according to any one of claims 1 to 8 is implemented.

10. A computer-readable medium storing a computer program, characterized in that: When the computer program is executed by a processor, the in-vehicle audio control method according to any one of claims 1 to 8 is implemented.

Images