Intelligent sound equipment sound field adjusting method and intelligent sound equipment system

Abstract

The embodiment of the invention provides an intelligent sound equipment sound field adjusting method and an intelligent sound equipment system, and the system comprises a loudspeaker device which is configured to play audio data; the positioning device is configured to obtain loudspeaker position data and user position data; the loudspeaking device and the positioning device are deployed in a target space, and the loudspeaking device and the positioning device are in communication connection with the processor; the processor is configured to obtain loudspeaker position data and user position data through the positioning device; and determining a target sound field parameter of the loudspeaker device based on the loudspeaker position data and the user position data.

Description

Technical Field

[0001] The present invention relates to the field of audio technology, and in particular to a smart audio sound field adjustment method and a smart audio system. Background Art

[0002] As consumers' demand for audio experience increases, the development of smart speakers becomes increasingly important. Ultra-Wide Band (UWB) technology can be applied to the audio field to improve the sound field effect of the speakers due to its high-precision positioning capability and short-distance high-speed data transmission characteristics.

[0003] Since the listener's position and the room's acoustic environment also have a significant impact on the sound field effect of the sound system, in many practical application scenarios, in order to improve the sound effect, the listener still needs to manually adjust the speaker position, volume, etc., or rely on limited preset modes to determine the speaker orientation and mode. In addition, fixed sound field settings will also lead to an unsatisfactory listening experience for the audience.

[0004] Based on this, we hope to propose a smart audio system that can adapt to changing listening environments and intelligently improve the user's listening experience. Summary of the invention

[0005] One or more embodiments of the present specification provide a smart audio system. The smart audio system includes: a speaker configured to play audio data; a positioning device configured to obtain speaker position data and user position data; the speaker and the positioning device are deployed in the target space, and the speaker and the positioning device are communicatively connected to a processor; the processor is configured to obtain the speaker position data and the user position data through the positioning device; and based on the speaker position data and the user position data, determine the target sound field parameters of the speaker.

[0006] One or more embodiments of the present specification provide a smart audio sound field adjustment method, which is executed by a processor and includes: obtaining speaker position data and user position data through a positioning device; and determining target sound field parameters of the speaker device based on the speaker position data and the user position data.

[0007] One or more embodiments of the present specification provide a smart audio sound field adjustment device, including a processor, wherein the processor is used to execute the above-mentioned smart audio sound field adjustment method.

[0008] One or more embodiments of the present specification provide a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the above-mentioned smart audio sound field adjustment method. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0010] Figure 1 is a schematic diagram of application scenarios of the smart audio system according to some embodiments of this specification;

[0011] Figure 2 is an exemplary flow chart of a sound field adjustment method according to some embodiments of this specification;

[0012] Figure 3 is an exemplary flow chart of determining target sound field parameters according to some embodiments of this specification;

[0013] Figure 4 is an exemplary schematic diagram of a listening prediction model according to some embodiments of this specification;

[0014] Figure 5 is another exemplary flow chart for determining target sound field parameters according to some embodiments of this specification;

[0015] Figure 6 is an exemplary schematic diagram of a noise prediction model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0016] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0017] It should be understood that the "system", "device", "unit" and / or "equipment" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0018] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0019] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0020] Figure 1 It is a schematic diagram of application scenarios of the smart audio system shown in some embodiments of this specification.

[0021] The application scenarios of the smart audio system shown in some embodiments of this specification may include smart audio sound field adjustment in various scenarios such as conference rooms, living rooms, stages, and cinemas. For example, according to the user's position and activities in the scene, the smart audio system adjusts the sound effects in real time to provide users with a good listening experience.

[0022] In some embodiments, the application scenario may include a speaker 110, a processor 120, etc. In some embodiments, the speaker 110 and the processor 120 may be deployed in a target space. The target space refers to a space where audio needs to be played. In some embodiments, the target space may include multiple types. For example, it may be a living room, a conference room, etc.

[0023] In some embodiments, the components in the smart audio system can be connected and / or communicated with each other via a network (e.g., wireless connection, wired connection, or a combination thereof). For example, the speaker device 110 can be connected to the processor 120 via a network.

[0024] The speaker 110 refers to a hardware device for converting an electrical signal into an acoustic signal. The speaker can play audio data. The audio data is data related to the audio that the user wants to play. For example, music, broadcasts, game sound effects, etc. can be played.

[0025] In some embodiments, the speaker device 110 may include at least two speakers (e.g., a left speaker 110-1 and a right speaker 110-2). The speaker device 110 may be a tweeter, a woofer, a floor-standing speaker, a wall-mounted speaker, or the like, or any combination thereof. The at least two speakers may be the same or different. In some embodiments, the at least two speakers may have different channels. For example, the channels of the at least two speakers may be a left channel and a right channel, respectively.

[0026] The processor 120 is used to process information and / or data related to the application scenario of the smart audio system.

[0027] The processor 120 may be a separate device, or may be integrated or partially integrated into other devices in the application scenario. For example, the processor may be disposed inside the speaker device 110 .

[0028] In some embodiments, the application scenario may further include a positioning device (not shown in the figure). The positioning device may be deployed in the target space. In some embodiments, the positioning device may include at least one ultra-wideband sensor. The ultra-wideband sensor may be installed on the speaker 110. For example, an ultra-wideband sensor may be installed on each speaker. In some embodiments, the ultra-wideband sensor may be configured on a smart terminal. The positioning device is communicatively connected to the processor 120.

[0029] In some embodiments, the processor may process data, information and / or processing results obtained from other devices in the application scenario, and execute program instructions based on these data, information and / or processing results to perform one or more functions described in this specification.

[0030] The above description is for illustrative purposes only, and actual application scenarios may vary.

[0031] Figure 2 is an exemplary flow chart of a sound field adjustment method according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps 210 and 220. In some embodiments, the process 200 may be executed by the processor 120.

[0032] Step 210: Acquire speaker position data and user position data through a positioning device.

[0033] Speaker position data refers to parameters related to the position of the speaker. For example, the speaker position data can be expressed as the coordinates of the speaker position in a two-dimensional rectangular coordinate system or a three-dimensional rectangular coordinate system. For more information about speakers, see Figure 1 Related instructions.

[0034] User position data refers to the position of the user in the target space. For example, the user position data can be expressed as the coordinates of the position of the user's head in a three-dimensional rectangular coordinate system or a two-dimensional rectangular coordinate system. In some embodiments, the processor can use any point in the target space as the coordinate origin O, two axes Ox and Oy parallel to the horizontal plane, and an axis Oz perpendicular to the horizontal plane as coordinate axes to construct a three-dimensional rectangular coordinate system, or use any point in the target space as the coordinate origin O, two axes Ox and Oy parallel to the horizontal plane as coordinate axes to construct a two-dimensional rectangular coordinate system.

[0035] In some embodiments, the processor can obtain the speaker position data and the user position data by measuring the ultra-wideband sensor. For more information about ultra-wideband sensors, see Figure 1 Related instructions.

[0036] In some embodiments, the processor may also obtain the speaker position data and the user position data through other distance sensors, including but not limited to ultrasonic sensors, infrared sensors, radar sensors, etc.

[0037] In some embodiments, the processor may acquire the user location data by locating the position of the user's smart terminal (eg, a mobile phone) based on an ultra-wideband tag configured on the user's smart terminal.

[0038] UWB tags are small devices carried by users or objects. UWB tags can actively transmit UWB signals or passively respond to UWB interrogation signals. UWB sensors are responsible for receiving UWB signals, which can determine the location of the smart terminal, i.e., the user location data.

[0039] In some embodiments, the processor may also obtain user location data in other feasible ways. For example, based on one or more ultra-wideband sensors, the processor may obtain user location data based on RSSI (Received Signal Strength Indicator) positioning method or AOA (Angle of Arrival) positioning method.

[0040] Step 220: Determine the target sound field parameters of the speaker device based on the speaker position data and the user position data.

[0041] In some embodiments, the processor may determine the target sound field parameters in a variety of ways. For example, the processor may calculate the distance between the user and the plurality of speakers based on the speaker position data and the user position data, and determine the target sound field parameters based on the distance. The loudness of the speaker is positively correlated with the distance, and the speaker delay parameter is negatively correlated with the distance.

[0042] For example, taking the confirmation of loudness and speaker delay parameters in the sound field parameters as an example, when there are two speakers in the target space, in order to ensure that the user has a good listening experience, the parameters of the sound waves emitted by different speakers need to be close when they reach the user's location. Therefore, the loudness of the speaker farther away from the user's location can be increased and its delay time can be reduced, so that the loudness and delay time of the sound waves emitted by the two speakers at the user's location are close.

[0043] In some embodiments, the processor may also determine the target sound field parameters in other ways. For more information, see Figure 3 , Figure 5 Related instructions.

[0044] In some embodiments, the speaker position data includes the position and angle of the speaker, and the target sound field parameters further include the speaker target position.

[0045] The processor can obtain the speaker angle in a variety of ways. In some embodiments, the processor can obtain the speaker angle by reading the steering parameter of the drive device. For example, when the speaker is fixedly connected to the drive device, with the wall as a reference, the initial angle of the speaker is 90°, and the drive device rotates 30° clockwise, then the speaker angle is 120°. The steering parameter of the drive device is the angle of rotation of the drive device.

[0046] In some embodiments, the processor may also obtain the speaker angle based on an angle sensor measurement, where the angle sensor includes but is not limited to a Hall effect angle sensor, a magnetoresistive effect angle sensor, and the like.

[0047] The speaker target position refers to the position to which the speaker needs to be moved. For example, the speaker target position includes the spatial coordinates to which the speaker needs to be moved and the angle to which the speaker needs to be rotated.

[0048] In some embodiments, the processor may determine the target position of the speaker based on the speaker position data and the user position data. For example, the speaker may be moved to within a preset distance range of the user's position, the speaker may be oriented toward the user, etc. The preset distance range is the range of the distance between the speaker and the user, which may be manually preset.

[0049] In some embodiments of the present specification, by considering the influence of the position and angle of the speaker on the user's auditory experience, the target sound field parameters obtained are more accurate.

[0050] In some embodiments, the processor may calculate the distance the speaker needs to move and the angle of rotation based on the current position of the speaker and the target position of the speaker, and drive the speaker to move to the target position of the speaker through a driving device.

[0051] The drive unit is the device used to move the speaker unit.

[0052] In some embodiments, the driving device may include at least one driving device. The driving device may be installed on the speaker. The number and installation position of the driving device may be set as needed, for example, the driving device may be installed only at the bottom of the main speaker; for another example, the driving device may be installed at the bottom of each speaker.

[0053] The driving device can drive the corresponding loudspeaker to move to a target loudspeaker position. For example, the loudspeaker can be driven to move a target distance and / or rotate a target angle.

[0054] The structure of the driving device can be motor drive, gear transmission, linear drive, etc. or any combination thereof. For example, the driving device may include a motor and a gear transmission structure, the motor is connected to the bottom of the speaker through the gear transmission structure, and the motor amplifies the rotational power through the gear transmission structure and transmits it to the speaker to realize its rotation.

[0055] The driving device may be communicatively connected to the processor. For example, the processor may transmit the target sound field parameter to the driving device, and the driving device drives the speaker to move based on the target sound field parameter.

[0056] In some embodiments of the present specification, a speaker is moved to a target speaker position by a driving device, thereby eliminating the adverse effects of inappropriate speaker position and sound angle on the user's auditory perception and improving the user experience.

[0057] In some embodiments, process 200 may further include step 230 .

[0058] Step 230 : Based on the target sound field parameters, generate a sound field parameter adjustment instruction and send it to the speaker device to adjust the sound field parameters of the speaker device.

[0059] The sound field parameter adjustment instruction is an instruction for the user to adjust the sound field parameters of the speaker device. For example, the sound field parameter adjustment instruction may be to adjust the loudness of the speaker to 40 dB.

[0060] In some embodiments, the processor may determine the adjustment amount of the sound field parameter adjustment instruction based on the target sound field parameter, and send the sound field parameter adjustment instruction to the speaker device, which adjusts the sound field parameter to the target sound field parameter in response to the sound field parameter adjustment instruction.

[0061] In some embodiments of the present specification, target sound field parameters are determined based on speaker position data and user position data, which can ensure a good sound field at the user's location and improve user experience.

[0062] Figure 3FIG. 1 is an exemplary flow chart of determining target sound field parameters according to some embodiments of this specification. Figure 3 As shown, the process 300 includes the following steps. In some embodiments, the process 300 may be executed by the processor 120 .

[0063] Step 310: Determine spatial listening data based on the candidate sound field parameters, the spatial structure data, and the speaker position data. Figure 2 Related instructions.

[0064] Candidate sound field parameters are sound field parameters that may be determined as target sound field parameters. For more information about sound field parameters, see Figure 2 Related instructions.

[0065] In some embodiments, the processor may obtain candidate sound field parameters in a variety of ways. For example, the processor may uniformly select values ​​at preset value intervals within a preset value range based on the current sound field parameters of the speaker, and combine to obtain multiple groups of candidate sound field parameters. The preset value range includes the current sound field parameters of the speaker, and the preset value ranges corresponding to different sound field parameters may be different. The preset value range and the preset value interval may be set manually.

[0066] In some embodiments, the processor may also determine multiple groups of candidate sound field parameters by constructing pre-selected feature vectors and matching them in a vector database based on the pre-selected feature vectors. Figure 4 Related instructions.

[0067] Spatial structure data refers to data related to the structure of the target space. For example, spatial structure data may include the shape, size, ceiling height, etc. of the target space.

[0068] In some embodiments, the processor may determine the spatial structure data by obtaining user input. The spatial structure data may also be determined in other feasible ways, for example, by calling system preset data.

[0069] Spatial listening data refers to the listening quality data at each position in the target space when playing audio using different candidate sound field parameters.

[0070] Listening quality data refers to data related to audio playback quality, which is related to the frequency, loudness, and noise level of the target space of the audio signal. For example, the closer the frequency and loudness are to the standard value and the smaller the noise is, the higher the listening quality data is. The standard value can be preset manually.

[0071] In some embodiments, the processor can use a spectrum analyzer to analyze the spectrum of the audio signal to determine the noise level. Exemplarily, if there is an obvious peak in the spectrum, it means that there is single-tone noise in the audio signal. At the same time, there is broadband noise in the audio signal. The processor can determine the noise level based on the size of the area occupied by the obvious peak and the fluctuation of the sound pressure. The larger the area occupied by the obvious peak and the greater the fluctuation of the sound pressure, the greater the noise. The processor can measure the frequency and loudness of the audio signal through a frequency meter and a loudness meter.

[0072] The frequency, loudness, and noise level of the target space of the audio signal may also be determined by a sound sensor or other means.

[0073] In some embodiments, the processor may determine the spatial listening data by querying a first preset table based on the candidate sound field parameters, the spatial structure data, and the speaker position data, wherein the first preset table includes a correspondence between the candidate sound field parameters, the spatial structure data, the speaker position data, and the spatial listening data, which may be determined based on historical data.

[0074] In some embodiments, the processor may also determine the spatial listening data using a listening prediction model. For more information, see Figure 4 Related instructions.

[0075] Step 320: Determine preferred listening position data based on the spatial listening data.

[0076] The preferred listening position data refers to the position with the highest listening quality data in the spatial listening data corresponding to each set of candidate sound field parameters.

[0077] In some embodiments, the processor may determine the position with the highest listening quality data in the spatial listening data as the preferred listening position data.

[0078] Step 330: Determine target sound field parameters based on the preferred listening position data and the user position data.

[0079] The processor may determine the target sound field parameters based on a variety of feasible methods. In some embodiments, the processor may calculate the distance between the preferred listening position data and the user position data, and determine the target sound field parameters based on the distance.

[0080] For example, the processor can sort the calculated distances between the preferred listening position data and the user position data according to the distance size, and determine whether the minimum distance is less than a first distance threshold. In response to a yes judgment, the processor determines the candidate sound field parameters corresponding to the minimum distance as the target sound field parameters; otherwise, the processor regenerates multiple groups of candidate sound field parameters until there are candidate sound field parameters corresponding to the listening position data and the user position data whose distance is less than the first distance threshold.

[0081] The first distance threshold can be manually set. For example, the processor can regenerate multiple sets of candidate sound field parameters by increasing the preset value range and reducing the preset value interval. For another example, the processor can regenerate multiple sets of candidate sound field parameters by increasing the distance threshold. For more information about the distance threshold, see Figure 4 Related instructions.

[0082] In some embodiments, the processor may also directly determine the candidate sound field parameter corresponding to the minimum distance as the target sound field parameter.

[0083] In some embodiments, in response to the number of users in the target space satisfying a preset number condition, the processor may determine the target sound field parameters based on a set of user position data, preferably listening position data, of the target space, wherein the preset number condition is that the number of users is greater than one.

[0084] The user position data set refers to a set of user position data of multiple users. It can be understood that for different users, the distance between the preferred listening position data and the user position data is different.

[0085] In some embodiments, the processor may calculate the sum of distances between the preferred listening position data and multiple user position data in the user position data set, and sort them according to the size of the sum of distances to obtain the minimum sum of distances, and determine the candidate sound field parameters of the preferred listening position data corresponding to the minimum sum of distances as the target sound field parameters.

[0086] In some embodiments of this specification, when there are multiple users, the target sound field parameters obtained can bring a good experience to as many users as possible, taking into account the overall listening experience of the multiple users.

[0087] In some embodiments of the present specification, target sound field parameters are determined based on multiple groups of candidate sound field parameters, while taking into account the influence of spatial structure data and speaker position data on sound propagation and reflection, so that the obtained target sound field parameters are more accurate.

[0088] It should be noted that the above description of process 300 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0089] Figure 4 is an exemplary schematic diagram of a listening prediction model according to some embodiments of this specification.

[0090] like Figure 4As shown, in some embodiments, the processor 120 may determine the spatial listening data 450 through the listening prediction model 440 based on the candidate sound field parameters 410, the spatial structure data 420, and the speaker position data 430. For more information about the speaker position data, see Figure 2 For more information about candidate sound field parameters, spatial structure data, and spatial listening data, see Figure 3 Related instructions.

[0091] The listening prediction model is a model used to predict spatial listening data corresponding to the candidate sound field parameters. In some embodiments, the listening prediction model may be a machine learning model, such as a neural network (NN) model.

[0092] In some embodiments, the listening prediction model can be obtained through training based on a large number of first training samples with a first label. In some embodiments, the training process may include: inputting the first training sample into the initial listening prediction model, substituting the output of the initial listening prediction model and the first label into the formula of a predefined loss function, and calculating the value of the loss function; according to the value of the loss function, reversely updating the model parameters in the initial listening prediction model; this step can be performed using various methods. For example, the update can be based on the gradient descent method. When the iteration end condition is met, the iteration ends and the trained listening prediction model is obtained. The iteration end condition includes but is not limited to the convergence of the loss function and the number of iterations reaching a threshold.

[0093] The first training sample includes sample sound field parameters, sample space structure data and sample speaker position data, and the first training sample can be acquired based on historical data.

[0094] The first label is the spatial listening data of the sample space corresponding to the first training sample. In some embodiments, multiple sound sensors can be installed in the sample space to record the sound data of each position after playing the audio with the sample sound field parameters, and the sound data is analyzed to obtain the spatial listening data as the first label.

[0095] In some embodiments, the processor may also obtain environmental characteristic data of the target space; determine candidate sound field parameters based on the environmental characteristic data and the spatial structure data; and determine spatial listening data 450 through a listening prediction model 440 based on candidate sound field parameters 410, spatial structure data 420, speaker position data 430, environmental characteristic data 460-1, spatial object data 460-2, and predicted ambient noise data 460-3.

[0096] Predicted environmental noise data refers to the predicted environmental noise data at each location in the target space at a preset future time. For more information about predicted environmental noise data, see Figure 6Related instructions.

[0097] Environmental characteristic data refers to data related to the environment of the target space. For example, the environmental characteristic data includes the temperature, humidity, etc. of the target space.

[0098] In some embodiments, the processor may obtain environmental characteristic data by measuring temperature sensors, humidity sensors, etc. deployed in the target space.

[0099] In some embodiments, the processor may construct a space-corresponding feature vector based on environmental feature data, space structure data, and speaker device data corresponding to a certain space.

[0100] A feature vector is a vector constructed based on the environmental feature data, spatial structure data, and speaker device data corresponding to the space. There are many ways to construct a feature vector. For example, a feature vector can be (x, y, z), where x represents the environmental feature data, y represents the spatial structure data, and z represents the speaker device data.

[0101] The vector database includes multiple reference vectors and reference sound field parameters corresponding to each of the multiple reference vectors.

[0102] The reference vector is constructed based on the historical environmental feature data, historical spatial structure data, and historical speaker device data in the historical data. The reference sound field parameters corresponding to the reference vector are the appropriate historical sound field parameters used by the speaker device in the historical space corresponding to the reference vector, which can be obtained by reading the device log of the speaker or by other feasible methods. The vector to be matched is constructed based on the environmental feature data, spatial structure data, and speaker device data corresponding to the target space. The construction method of the reference vector and the vector to be matched is shown in the feature vector above.

[0103] In some embodiments, the processor may calculate the distance between the reference vector and the vector to be matched, and determine the candidate sound field parameters corresponding to the vector to be matched. For example, the processor may use a reference vector whose distance from the vector to be matched is less than a distance threshold as a target vector, and determine the reference sound field parameters corresponding to the target vector as candidate sound field parameters.

[0104] The spatial object data refers to data related to an object in the target space. For example, the spatial object data includes the size, position, etc. of the object in the target space.

[0105] In some embodiments, the processor may determine the spatial object data by a variety of methods. For example, the processor may obtain the spatial object data by using parameters such as the object position of the target space uploaded by the user. The user may modify and update the parameters such as the object position uploaded by the user at any time. For another example, the processor may determine the spatial object data by using methods such as an object detection algorithm and a depth sensing algorithm based on the image data of multiple angles in the target space directly uploaded by the user.

[0106] In some embodiments, when the input of the listening prediction model includes environmental feature data, spatial object data and predicted environmental noise data, the first training sample also includes sample environmental feature data, sample spatial object data and sample environmental noise data corresponding to the sample space, which can be obtained based on historical data.

[0107] In some embodiments of the present specification, the spatial listening data obtained is more accurate by taking into account the influence of environmental characteristics and objects in the target space on the propagation and reflection of sound.

[0108] In some embodiments, the processor may also determine sound feature data based on sound sensor data; based on candidate sound field parameters 410, spatial structure data 420, speaker position data 430, sound feature data 470-1, and feature reference degree 470-2, determine spatial listening data 450 through a listening prediction model 440.

[0109] For more information about candidate sound field parameters, see Figure 3 Related instructions.

[0110] Sound sensor data refers to the sound data collected by the sound sensor.

[0111] In some embodiments, the sound collection device can transmit the sound sensing data to the processor in the form of electrical signals.

[0112] The sound collection device refers to a hardware device used to collect sound data of the target space. In some embodiments, the sound collection device may include at least one sound sensor. The sound sensor may include various types of microphones, sound level meters, sound spectrometers, etc. or any combination thereof.

[0113] At least one sound sensor may be deployed in the target space. The number of sound sensors and the positions where they are deployed in the target space may be set as required.

[0114] The sound collection device may be in communication with the processor. For example, the processor may obtain sound sensing data based on the sound collection device, and determine sound feature data based on the sound sensing data.

[0115] Sound feature data refers to features of sound data, for example, sound feature data includes loudness data, spectrogram, sound intensity data, etc.

[0116] In some embodiments, the processor may analyze and determine the sound feature data based on the sound sensor data. For example, the processor may analyze the sound sensor data to obtain loudness data; perform short-time Fourier transform on the sound sensor data to obtain a spectrum image, construct a spectrogram based on the time domain image and spectrum image of the sound sensor data; obtain sound intensity data by measuring with a sound intensity meter. Sound feature data, or determine the sound feature data by other feasible methods.

[0117] The feature reference level refers to the reference level of the sound data collected by the sound sensor when determining the sound feature data. In some embodiments, the feature reference level can be represented by a number between 0 and 1, and the larger the number, the higher the feature reference level. If the feature reference level corresponding to a sound sensor is 0, then the sound sensor does not need to be considered when determining the sound feature data.

[0118] The characteristic reference degree is negatively correlated with the distance between the sound sensor and the speaker. In some embodiments, the processor can determine the characteristic reference degree by querying a second preset table. The second preset table includes a correspondence between the distance between the sound sensor and the speaker and the characteristic reference degree, which can be set based on experience.

[0119] In some embodiments, the characteristic reference degree is also positively correlated with the measurement accuracy of the sound sensor, and the lower the measurement accuracy of the sound sensor, the lower the corresponding characteristic reference degree. The measurement accuracy of the sound sensor can be determined based on historical data.

[0120] In some embodiments, when the input of the listening prediction model includes sound feature data and feature reference levels, the first training sample also includes sample sound feature data and sample feature reference levels, which may be acquired based on historical data.

[0121] In some embodiments of the present specification, by performing weighted processing on different sound sensors when determining sound feature data, the obtained sound feature data is more accurate, and the spatial listening data is determined based on the sound feature data, so that the obtained spatial listening data is more accurate.

[0122] In some embodiments of the present specification, a listening prediction model is used to predict spatial listening data, so that the obtained spatial listening data is more accurate, and target sound field parameters are determined based on the spatial listening data, so that the obtained target sound field parameters better meet user needs and improve user experience.

[0123] Figure 5 FIG. 1 is another exemplary flow chart for determining target sound field parameters according to some embodiments of this specification. Figure 5As shown, the process 500 includes the following steps. In some embodiments, the process 500 may be executed by a processor.

[0124] Step 510: Determine a preferred listening position range based on the spatial listening data and preset parameters.

[0125] For more information about spatial listening data, see Figure 3 .

[0126] The preset parameters refer to preset parameter standards and conditions used to determine the preferred listening position range. In some embodiments, the preset parameters may be a preset radius and / or a preferred threshold.

[0127] The preset radius refers to a pre-set radius value reflecting the preferred listening position. For example, with the speaker as the center, the preset radius is 2 meters.

[0128] The preferred threshold refers to a preset critical value of the listening quality data, for example, the preferred threshold is 80 points.

[0129] In some embodiments, the preset radius and the preferred threshold can be determined based on historical experience, experimental data, etc.

[0130] In some embodiments, the processor can adjust the preset radius and the preferred threshold based on the current number of users and the spatial structure data. For example, the more current users there are and the larger the spatial area is, the larger the preset radius is set, or the smaller the preferred threshold is set.

[0131] The current number of users refers to the number of users in the target space. For more information about spatial structure data, see Figure 3 .

[0132] The preferred listening position range refers to the range of the area with high listening quality. In some embodiments, the preferred listening position range can be represented by geometric shapes, coordinates, etc. For example, the preferred listening position range is a circular area with a radius of 1 meter and the center position of the two speakers as the center.

[0133] In some embodiments, the processor may determine the preferred listening position range in a variety of ways based on the spatial listening data and preset parameters.

[0134] In some embodiments, the processor may determine the preferred listening position range based on the spatial listening data and a preset radius. For example, if the preset radius is set to 2 meters, the processor may determine a circular area with a radius of 2 meters as the preferred listening position range based on the spatial listening data and with the preferred listening position as the center.

[0135] It is understandable that the sound field of the target space may be non-uniform. For example, two users are located at a position close to the speaker and a position far away from the speaker respectively, and the two users have different listening experiences.

[0136] In some embodiments, the processor may also determine a preferred listening position range based on the spatial listening data and the preferred threshold. Exemplarily, the processor may use the region formed by the set of positions where the listening quality data in the spatial listening data is greater than the preferred threshold as the preferred listening position range. The center of the region is the position with the best listening quality; the boundary of the region is the position of the critical point below the preferred threshold; the listening quality outside the region starts to decrease from the boundary.

[0137] Step 520: Determine target sound field parameters based on the preferred listening position range and user position data.

[0138] For more information about user position data and target sound field parameters, see Figure 2 Related instructions.

[0139] In some embodiments, the processor may determine the target sound field parameters in a variety of ways based on the preferred listening position range and the user position data.

[0140] In some embodiments, the processor may determine a corresponding preferred listening position range based on each set of candidate sound field parameters, and determine whether the range includes the user position data of the current user.

[0141] If the user position data of the current user is not included, the processor may regenerate at least one set of candidate sound field parameters. For example, assuming that the current user is located on the right side of the room, and the preferred listening position range of the candidate sound field parameters is concentrated on the left side, the processor will regenerate at least one set of candidate sound field parameters.

[0142] If the user position data of the current user is included, the processor may determine the group with the largest preferred listening position range among all candidate sound field parameters including the user position data of the current user as the target sound field parameters. For example, assuming that there are three groups of candidate sound field parameters corresponding to preferred listening position ranges including the user position data of the current user, the preferred listening position range of the first group is 10 square meters, the second group is 8 square meters, and the third group is 12 square meters, then the processor uses the first group of candidate sound field parameters with a range of 12 square meters as the target sound field parameters finally determined.

[0143] Candidate sound field parameters and their generation methods refer to Figure 3 Related instructions.

[0144] In some embodiments, step 520 may further include steps 521 to 523 described below.

[0145] Step 521 : determining the user centrality corresponding to the candidate sound field parameters based on the preferred listening position range corresponding to the candidate sound field parameters and the user position data.

[0146] User centrality refers to an indicator that measures the position of user location data within the preferred listening position range. In some embodiments, user centrality can be represented by distance, score, etc. For example, user centrality is "high", "medium", or "low".

[0147] In some embodiments, the processor may determine the user centrality in a variety of ways based on the preferred listening position range corresponding to the candidate sound field parameters and the user position data.

[0148] In some embodiments, the user centrality may be positively correlated to the distance between the user position and the center of the preferred listening position range. For example, the processor may construct a polygon based on the outline of the preferred listening position range, calculate the distance from the user position to the center of mass of the polygon, and use the inverse of the distance as the user centrality. The closer the user position is to the center of mass, the higher the corresponding user centrality.

[0149] The range listening parameter refers to the average value of the listening quality data at each position within the preferred listening position range.

[0150] For more information on listening quality data, see Figure 3 Related instructions.

[0151] Step 522: Determine the preferred index based on the user centrality and the range listening parameters.

[0152] The preference index of the candidate sound field parameter is a quantitative index to measure the suitability of the candidate sound field parameter to be selected as the target sound field parameter. In some embodiments, the preference index can be expressed as a numerical value, etc. For example, the preference index is 0.5.

[0153] It can be understood that a preferred index is determined corresponding to each group of candidate sound field parameters, and the target sound field parameters can be selected by comparing the preferred indexes of the candidate sound field parameters of each group.

[0154] In some embodiments, the processor may determine the preference index in a variety of ways based on user centrality, range listening parameters, etc.

[0155] In some embodiments, the preference index may be positively correlated to the size of the preferred listening position range, the user centrality, and the range listening parameter.

[0156] For example, the preference index may be determined by Formula 1.

[0157] I = C1 × A + C2 × D + C3 × Q (1)

[0158] Among them, I represents the preferred index, C1, C2, and C3 represent weight coefficients, A represents the size of the preferred listening position range, D represents the user centrality, and Q represents the range listening parameter.

[0159] The weight coefficient can be determined based on manual preset.

[0160] Step 523: Determine the target sound field parameters based on the optimization index.

[0161] In some embodiments, the processor may determine the target sound field parameters in a variety of ways based on the preference index. For example, the processor may use a group of candidate sound field parameters with the largest preference index within the preferred listening position range as the target sound field parameters.

[0162] Some embodiments of this specification can more accurately optimize the target sound field parameters by comprehensively considering multiple factors such as user centrality, thereby ensuring that the smart audio system can provide a better listening experience.

[0163] In some embodiments, the processor may also determine the target sound field parameters based on the user position data set and the preferred listening position range of the target space in response to the number of users in the target space satisfying a preset number condition.

[0164] For more information about the preset quantity conditions and how to determine the target sound field parameters, see Figure 3 Related instructions.

[0165] In some embodiments, in response to the number of users in the target space satisfying a preset number condition, the processor may determine the target sound field parameters in a variety of ways based on a user position data set, a preferred listening position range, and a preferred index.

[0166] For example, the processor may use a group of candidate sound field parameters whose preferred listening position range includes all current user position data and whose priority index is the largest among the candidate sound field parameters as target sound field parameters.

[0167] The method of determining the priority index is similar to the method of determining the priority index in step 522. For detailed description, please refer to the relevant description of the above step 522. In some embodiments, when determining the priority index by formula 1, the size of the weight coefficient C1 may be positively correlated with the number of users.

[0168] In some embodiments, when the preference index is determined by Formula 1, the user centrality D may be the average of all user centralities in the candidate sound field parameters.

[0169] Some embodiments of this specification can adapt to different scenario requirements by considering the number of users in the target space, scientifically determine the target sound field parameters, and ensure that the smart audio system can provide a better listening experience.

[0170] Some embodiments of this specification can effectively improve the accuracy of sound field adjustment by combining spatial listening data, preset parameters, and user position data to determine target sound field parameters, ensuring that the user is within the preferred listening position range; they can also dynamically adjust the sound field parameters to adapt to changes in the user's position, thereby providing a consistent and high-quality listening experience in different environments and usage scenarios. In addition, by calculating and adjusting the preferred listening position range, it is possible to better adapt to spatial structure and environmental characteristics, and improve the applicability and intelligence level of the smart audio system.

[0171] In some embodiments, the processor can determine the predicted environmental noise data through a noise prediction model based on historical environmental noise data, environmental characteristic data, future weather data, spatial structure data, and spatial object data; and determine the target sound field parameters based on the predicted environmental noise data, user position data, and speaker position data.

[0172] Future weather data refers to data reflecting weather conditions within a preset future time. The preset future time can be a time period from the current time point to a future time point. For example, 16:00 on the current day to 16:00 on the next day.

[0173] In some embodiments, the future weather data may be at least one of temperature, humidity, rainfall probability, wind speed, etc. In some embodiments, the future weather data may be represented in a table, a tuple, etc. For example, the future weather data is {(24 hours), (rainfall probability 30%), (northeast wind), (sunny)}.

[0174] In some embodiments, future weather data may be obtained based on the National Meteorological Service website, weather forecasts, and the like.

[0175] Predicted environmental noise data refers to the predicted environmental noise data of each location in the target space within a preset future time.

[0176] Environmental noise data refers to the noise data in the target space other than the audio played by the speakers, such as the operating noise of electrical appliances, animal calls, and sounds made by users.

[0177] In some embodiments, the environmental noise data may include noise loudness, frequency, etc. Noise loudness refers to the volume of the noise. Frequency refers to the vibration frequency of the noise.

[0178] In some embodiments, the environmental noise data may be acquired via sensors or the like.

[0179] Historical environmental noise data refers to environmental noise data in the target space in the past. In some embodiments, historical environmental data can be obtained based on a variety of methods. For example, it can be obtained based on historical data.

[0180] In some embodiments, the processor may determine predicted environmental noise data through a noise prediction model based on historical environmental noise data, environmental characteristic data, future weather data, spatial structure data, and spatial object data.

[0181] For descriptions of environmental characteristic data, spatial structure data, and spatial object data, see Figure 4 .

[0182] The noise prediction model can be a machine learning model.

[0183] In some embodiments, the noise prediction model may be a single-layer model or a multi-layer model.

[0184] In some embodiments, the noise prediction model may include a prediction layer.

[0185] In some embodiments, the prediction layer may be a recurrent neural network model (RNN) or other machine learning models.

[0186] In some embodiments, the prediction layer may output predicted environmental noise data based on historical environmental noise data, environmental characteristic data, future weather data, spatial structure data, and spatial object data.

[0187] In some embodiments, the prediction layer can be obtained by training multiple labeled second training samples. For example, multiple labeled second training samples can be input into the initial prediction layer, and a loss function can be constructed by using the labels and the results of the initial prediction layer. The parameters of the initial prediction layer can be iteratively updated based on the loss function by gradient descent or other methods. When the preset conditions are met, the prediction layer training is completed and a trained prediction layer is obtained. The preset conditions can be that the loss function converges, the number of iterations reaches a threshold, etc.

[0188] In some embodiments, the second training sample may include at least sample environmental noise data, sample spatial structure data, sample spatial object data, sample environmental feature data, and sample weather data of the second historical time at the first historical time. The first historical time is earlier than the second historical time.

[0189] The label may be the environmental noise data actually collected at the second historical time corresponding to the second training sample. The label may be determined based on historical experience or may be manually labeled.

[0190] The training process of the prediction layer is similar to that of the listening prediction model, which can be found in Figure 4 Description of the listening prediction model.

[0191] In some embodiments, the noise prediction model 620 may further include a separation layer 621 and a prediction layer 622 .

[0192] In some embodiments, the separation layer can be a Convolutional Neural Networks (CNN) model or other machine learning model.

[0193] like Figure 6 As shown, the separation layer 621 can determine the historical environmental noise data 614 based on the historical sound sensor data 611 and the historical playback audio data 612 .

[0194] The historical sound sensor data refers to the sound sensor data in the target space in the past time. The sound sensor data can be obtained by a sensor, etc. The historical sound sensor data can be determined based on the historical data.

[0195] The historical played audio data refers to the audio data played by the speaker device 110 in the target space in the past time. The historical played audio data can be acquired based on the historical data.

[0196] The historical environmental noise data 614 output by the separation layer 621 can be used as the input of the prediction layer 622 .

[0197] like Figure 6 As shown, the prediction layer 622 can determine the predicted environmental noise data 460-3 based on the historical environmental noise data 614, environmental characteristic data 460-1, future weather data 613, spatial structure data 420, and spatial object data 460-2 output by the separation layer.

[0198] For descriptions of environmental characteristic data, spatial structure data, and spatial object data, see Figure 4 Related instructions.

[0199] In some embodiments, the parameters of the separation layer can be obtained by training a noise prediction model. The noise prediction model is composed of a separation layer and a prediction layer.

[0200] In some embodiments, the third training sample of the noise prediction model includes sample sound sensor data of a first historical time, sample playback audio data, sample spatial structure data, sample spatial object data, sample environmental feature data of a second historical time, and sample weather data of a second historical time.

[0201] The label may be the actual collected environmental noise data corresponding to the third training sample. The label may be determined based on historical experience or may be manually labeled.

[0202] The sample sound sensor data and the sample playback audio data are input into the separation layer; the output of the separation layer is input into the prediction layer together with the sample environmental feature data, the sample weather data, the sample space structure data, and the sample space object data to obtain the predicted environmental noise data output by the prediction layer.

[0203] Based on the predicted environmental noise data output by the label and prediction layer, a loss function is constructed, and the parameters of the separation layer and the prediction layer are updated synchronously. Through parameter updating, the trained separation layer and prediction layer are obtained.

[0204] In some embodiments, when the trained separation layer and prediction layer meet preset conditions, the model training ends. The preset conditions may be that the loss function result converges or is less than a threshold value.

[0205] Obtaining the parameters of the separation layer through the above training method can, in some cases, help solve the problem of difficulty in obtaining labels when training the separation layer alone, and can also enable the separation layer to better obtain predicted environmental noise data that reflects the noise conditions of the target space.

[0206] Some embodiments of this specification, by adopting a noise prediction model that combines a separation layer and a prediction layer, can significantly improve the accuracy and stability of noise prediction, reduce the interference of environmental noise on sound quality, and enhance the overall system intelligence level and user satisfaction.

[0207] In some embodiments, the predicted ambient noise data may also be related to the sound characteristic data.

[0208] In some embodiments, the input of the noise prediction model may further include sound feature data, and the corresponding training samples may further include sample sound feature data.

[0209] For a description of the sound characteristics data of the processor, see Figure 4 For relevant instructions on the output of the noise prediction model for predicting environmental noise data, see Figure 6 Related instructions.

[0210] In some embodiments of the present specification, by considering sound feature data in noise prediction, the processor can more accurately determine and analyze the source of noise, and provide clearer and higher-quality audio effects; thereby improving the smart audio system's ability to adapt to and manage complex sound field environments, and improving the level of intelligence and user satisfaction.

[0211] like Figure 6 As shown, the processor may determine the target sound field parameters 640 based on the predicted environmental noise data 460 - 3 , the user position data 631 , and the speaker position data 430 .

[0212] In some embodiments, the processor may determine the target sound field parameters in various ways.

[0213] In some embodiments, the predicted environmental noise data may include predicted noise loudness. The predicted noise loudness refers to the volume of noise predicted in the future.

[0214] In some embodiments, the processor may determine the target sound field parameters based on the predicted noise loudness at the current user position, the user position data, and the speaker position data.

[0215] In some embodiments, the target sound field parameters may be positively correlated with the predicted noise loudness at the current user position. For example, based on the speaker position data, it is determined that the two speakers are located on the left and right sides of the central axis of the room, and based on the user position data, it is determined that the user is located on the left side of the central axis of the room. Based on this, the processor may preliminarily determine that the target sound field parameters are to increase the loudness of the speaker on the right side of the central axis of the room; and based on the fact that the predicted noise loudness of the current user is higher, the processor may finally determine that the target sound field parameters are to increase the loudness of the speakers on the left and right sides of the central axis of the room. In some embodiments, the predicted noise loudness may be the average of the noise loudness within the target area centered on the user. The target area may be a circular area with a preset radius centered on the user. The preset radius may be determined based on experimental data or manual preset.

[0216] In some embodiments, the predicted noise loudness may also be the average noise loudness of the entire target space.

[0217] In some embodiments, the processor may also input the predicted ambient noise data, as well as the candidate sound field parameters, spatial structure data, and speaker position data into a listening prediction model, and the listening prediction model may output spatial listening data; the processor may determine the preferred listening position data based on the spatial listening data; and the processor may determine the target sound field parameters based on the preferred listening position data and the user position data.

[0218] For detailed information about the listening prediction model and the method for determining the target sound field parameters, see Figure 4 For more information about predicting environmental noise data, see Figure 6 The aforementioned related instructions.

[0219] Some embodiments of this specification, by combining historical environmental noise data, environmental characteristic data and other factors, use noise prediction models to more accurately evaluate the noise conditions in the target space and optimize the sound field parameter settings of the speakers to ensure a better listening experience under various environmental conditions.

[0220] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0221] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0222] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0223] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0224] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0225] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0226] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A smart audio system, characterized in that: include: A speaker device configured to play audio data; A positioning device configured to obtain speaker position data and user position data; The speaker device and the positioning device are deployed in the target space, and the speaker device and the positioning device are communicatively connected with the processor; The processor is configured to: Acquiring the speaker position data and the user position data through the positioning device; and Based on the speaker position data and the user position data, target sound field parameters of the sound speaker are determined.

2. The system according to claim 1, characterized in that The processor is further configured to: Determining spatial listening data based on candidate sound field parameters, spatial structure data, and the speaker position data; Determining preferred listening position data based on the spatial listening data; as well as The target sound field parameters are determined based on the preferred listening position data and the user position data.

3. The system according to claim 2, characterized in that The preferred listening position data includes a preferred listening position range, and the processor is further configured to: Determining the preferred listening position range based on the spatial listening data and preset parameters; and The target sound field parameters are determined based on the preferred listening position range and the user position data.

4. The system according to claim 1, characterized in that The processor is further configured to: Based on historical environmental noise data, environmental characteristic data, future weather data, spatial structure data, and spatial object data, predicting environmental noise data through a noise prediction model; wherein the noise prediction model is a machine learning model; and The target sound field parameters are determined based on the predicted ambient noise data, the user position data, and the speaker position data.

5. A method for adjusting the sound field of an intelligent audio system, characterized in that: The method is executed by a processor, comprising: Acquire speaker position data and user position data through a positioning device; Based on the speaker position data and the user position data, a target sound field parameter of the sound speaker is determined.

6. The method according to claim 5, characterized in that The determining the target sound field parameters of the loudspeaker device based on the loudspeaker position data and the user position data comprises: Determining spatial listening data based on candidate sound field parameters, spatial structure data, and the speaker position data; Determining preferred listening position data based on the spatial listening data; and The target sound field parameters are determined based on the preferred listening position data and the user position data.

7. The method according to claim 6, characterized in that The preferred listening position data includes a preferred listening position range; The determining of the preferred listening position data based on the spatial listening data comprises: Determining the preferred listening position range based on the spatial listening data and preset parameters; and The determining the target sound field parameter based on the preferred listening position data and the user position data comprises: The target sound field parameters are determined based on the preferred listening position range and the user position data.

8. The method according to claim 5, characterized in that The determining the target sound field parameters of the loudspeaker device based on the loudspeaker position data and the user position data further comprises: Based on historical environmental noise data, environmental characteristic data, future weather data, spatial structure data, and spatial object data, predicting environmental noise data through a noise prediction model; wherein the noise prediction model is a machine learning model; and The target sound field parameters are determined based on the predicted ambient noise data, the user position data, and the speaker position data.

9. A smart audio sound field adjustment device, characterized in that: It includes a processor, and the processor is used to execute the smart audio sound field adjustment method according to any one of claims 5 to 8.

10. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the smart audio sound field adjustment method as described in any one of claims 5 to 8.

Images