Sleep aid system and method with dynamic natural sound simulation

Abstract

The application discloses a sleep auxiliary system and method for dynamic natural sound simulation, and particularly relates to the technical field of sleep assistance, which comprises obtaining historical preference data of a user and obtaining external environment data, generating a first reference coefficient based on the historical preference data of the user and the external environment data, wherein the historical preference data comprises a sound type, a volume value and a playing time length; obtaining historical sleep cycle data of the user, and inputting the first reference coefficient and the historical sleep cycle data into a pre-constructed first sound training model; the application determines a sleep state based on a real-time obtained physiological evaluation coefficient, compares the sleep state with a preset sleep time length, and optimizes a sound playing strategy, thereby realizing dynamic intelligent regulation and control of sleep assistance. In the sleep process, sound parameters can be adjusted in real time according to actual time length conditions of a light sleep state, a deep sleep state, a dream state or a wake-up state of the user.

Description

Technical Field

[0001] This invention relates to the field of sleep aid technology, and more specifically, to a sleep aid system and method based on dynamic natural sound simulation. Background Technology

[0002] Traditional sleep aids and wake-up devices are typically single-function, making it difficult to balance the simulation of natural sounds with a gentle wake-up experience. Sleep aids often help users fall asleep through white noise, soft music, or other fixed sound effects, but neglect the simulation of dynamic natural sounds. Wake-up devices, on the other hand, often wake users with abrupt ringing or vibrations, easily disrupting their sleep cycles and causing negative emotions and physical discomfort.

[0003] Existing methods, such as Chinese patent application CN101940813A, disclose a multifunctional sleep aid system and its control method. This system includes an ambient sound device and an air humidity regulator within the cabinet. The air humidity regulator contains an air quality regulator and an air purification and circulation device. The cabinet's surface integrates an ambient light device, a humidifier mist tube, a table lamp, the ambient light device, the ambient sound device, and a clock device. These devices are controlled manually and automatically, and the cabinet contains a computer control system. This method expands the functionality of the bedside table and humidifier, improving sleep quality. However, research and application of this method and existing technologies have revealed at least the following drawbacks:

[0004] The above methods simply control the switching on and off of various devices and adjust basic parameters, lacking an automatic intelligent control mechanism based on in-depth analysis of the sleep process, and cannot adapt well to the dynamic changes during sleep.

[0005] Therefore, there is an urgent need for a system capable of dynamically simulating natural sounds. This system could assist users in falling asleep through natural ambient sound effects and gradually play natural wake-up sounds within a user-set wake-up time range, thereby reducing discomfort caused by morning grumpiness and improving users' sleep quality and wake-up experience. To this end, this invention provides a sleep aid system and method for dynamically simulating natural sounds. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a sleep aid system and method for dynamic natural sound simulation to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] In a first aspect, the present invention provides a sleep aid method based on dynamic natural sound simulation, comprising:

[0009] Step 1: Obtain the user's historical preference data and external environment data, and generate a first reference coefficient based on the user's historical preference data and external environment data. The historical preference data includes sound type, volume value and playback duration.

[0010] Step 2: Obtain the user's historical sleep cycle data, and input the first reference coefficient and historical sleep cycle data into the pre-built first sound training model to obtain the user's initial sound playback data;

[0011] Step 3: Obtain the user's physiological assessment coefficient within time period T, and determine the user's sleep state within time period T+a based on the physiological assessment coefficient;

[0012] Step 4: Based on the user's sleep state during the T+a time period, obtain the corresponding sleep duration, compare the corresponding sleep duration with the preset sleep duration, optimize the initial sound playback data based on the comparison results, and obtain the sound control strategy.

[0013] Furthermore, the external environment data includes the temperature difference, humidity difference, and sound intensity of the user's location per unit time; the method for generating the first reference coefficient based on the user's historical preference data and the external environment data includes:

[0014] Step s1: Analyze the historical preference data to obtain the historical preference coefficient. ;

[0015] Step s2: Analyze the external environmental data to obtain the environmental assessment coefficient. ;

[0016] Step s3: Let the current time be t, and the historical preference coefficient sequence be { }, The historical preference coefficient at the current time t;

[0017] The environmental assessment coefficient sequence is { }, The environmental assessment coefficient is given at the current time t; the attenuation factor is... , 1;

[0018] Step s4: Calculate the weighted sum of historical preference coefficients r = 1, 2, ..., t;

[0019] And the weighted sum of the environmental assessment coefficients. ;

[0020] Step s5: Calculate the historical preference coefficients With environmental assessment coefficient After comprehensive processing, the first reference coefficient is obtained, and its calculation formula is as follows:

[0021]

[0022] In the formula, Indicates the first reference coefficient. and As a weighting factor, and .

[0023] Furthermore, the method for obtaining the historical preference coefficient includes:

[0024] Step a1: There are n different sound types, and each sound type corresponds to a technical score. , i = 1, 2, ..., n;

[0025] Step a2: Based on the user's selection of sound type frequency Perform weighted adjustments, the This represents the proportion of times the i-th sound type was selected out of the total number of selections.

[0026] Step a3: The volume value V is non-linearly mapped according to a logarithmic relationship. Let the minimum volume value be... The maximum volume value is Its mapping formula is:

[0027]

[0028] In the formula, This represents the volume value obtained after performing a non-linear mapping on the volume value V.

[0029] Step a4: The sound duration C is processed using a piecewise function, let... , , For different duration thresholds; when hour, ;when hour, ;when hour, ;when hour, , This refers to the playback duration.

[0030] Step a5: Select the sound type Selecting the frequency Volume value after non-linear mapping and playback duration Perform formulaic calculations to obtain historical preference coefficients;

[0031]

[0032] In the formula, This represents the historical preference coefficient.

[0033] Furthermore, the method for obtaining the environmental assessment coefficient includes:

[0034] Step b1: Temperature standard deviation The Gaussian function is used for processing, and the standard temperature range is set as []. , The real-time temperature is The calculation formula is:

[0035]

[0036] Step b1: Humidity standard deviation The Gaussian function is used for processing, and the standard humidity range is set as []. , The real-time humidity is The calculation formula is:

[0037]

[0038] Step b3: Standard deviation of sound intensity The process is performed using a power function, and the standard sound intensity range is set as []. , The real-time sound intensity is The calculation formula is:

[0039]

[0040] Step b4: Environmental Assessment Coefficient The calculation formula is:

[0041]

[0042] In the formula, Indicates the environmental assessment coefficient. , and As a weighting factor, and 1.

[0043] Furthermore, the historical sleep cycle data includes each sleep state and its corresponding sleep duration, wherein the sleep states include light sleep, deep sleep, dream state, and awake state; the initial sound playback data includes the initial music type, initial volume, and initial playback duration.

[0044] Methods for obtaining users' historical sleep cycle data include:

[0045] Step c1: Obtain historical EEG data, construct a time-domain EEG graph with time in the historical EEG data as the horizontal axis and EEG waves in the historical EEG data as the vertical axis. The EEG waves include theta waves, delta waves, beta waves and alpha waves.

[0046] Step c2: Divide the brainwave time-domain map according to the brainwaves to obtain a historical brainwave set, which includes Z historical brainwave waveforms, including theta wave waveforms, delta wave waveforms, beta wave waveforms and alpha wave waveforms;

[0047] Step c3: Extract the z-th historical EEG waveform from the historical EEG set, where z is a positive integer and the initial value of z is 1;

[0048] Step c4: Extract the corresponding standard EEG waveform, calculate the similarity between the historical EEG waveform and the standard EEG waveform. If the similarity between the historical EEG waveform and the standard EEG waveform is greater than or equal to the preset EEG similarity threshold, extract the corresponding sleep duration and jump to step c5; if the similarity between the historical EEG waveform and the standard EEG waveform is less than the preset EEG similarity threshold, jump directly to step c5.

[0049] Step c5: Let z = z + 1, and jump back to step c3;

[0050] Step c6: Repeat steps c3 to c5 until z=Z, then end the loop and obtain the corresponding sleep duration.

[0051] Furthermore, the method for constructing the first sound training model includes:

[0052] Historical sound training data is divided into a sound training set and a sound test set to construct a regression network model. The historical sound training data includes sound feature data and corresponding initial sound playback data. The sound feature data includes a first reference coefficient and historical sleep cycle data. The sound feature data in the sound training set is used as the input to the regression network model, and the initial sound playback data corresponding to the sound feature data in the sound training set is used as the output of the regression network model. The regression network model is trained to obtain an initial first regression network model with the training objective of minimizing the sum of prediction accuracies. The initial first regression network model is evaluated using the sound test set. If the sum of prediction accuracies is less than a preset accuracy threshold, the corresponding initial first regression network model is used as the first sound training model. If the sum of prediction accuracies is greater than or equal to the preset accuracy threshold, the initial first regression network model is trained again using the original sound training set until the test results meet the set threshold. The initial first regression network model can be a decision tree regression network model, a random forest regression network model, a support vector machine regression network model, a linear regression network model, or a neural network model.

[0053] Furthermore, the methods for obtaining the user's physiological assessment coefficient within time period T include:

[0054] Acquire real-time physiological representation data of users, including the absolute value of heart rate difference, the absolute value of respiratory rate difference, brain wave type, and the absolute value of muscle electrical activity frequency difference;

[0055] Real-time physiological characterization data are normalized to obtain physiological assessment coefficients;

[0056] Methods for determining a user's sleep state within the T+a time period based on physiological assessment coefficients include:

[0057] The physiological assessment coefficients are input into a pre-built state prediction model to obtain the user's sleep state during the T+a time period. The specific training process of the state prediction model includes:

[0058] P sets of physiological evaluation coefficients are collected in advance, where P is an integer greater than 1. The corresponding prediction results are set for the physiological evaluation coefficients. The prediction results include light sleep state, deep sleep state, dream state and awake state. Different numerical labels are set for light sleep state, deep sleep state, dream state and awake state. The numerical labels of the prediction results are marked as prediction labels. The physiological evaluation coefficients and the corresponding prediction labels are converted into a set of feature vectors.

[0059] Each set of feature vectors is used as input to the state prediction model. The state prediction model outputs a set of predicted labels corresponding to each set of physiological evaluation coefficients and uses the actual predicted labels corresponding to each set of physiological evaluation coefficients as the prediction target. The actual predicted labels are the pre-set predicted labels corresponding to the physiological evaluation coefficients. The training objective is to minimize the sum of prediction errors of the predicted labels corresponding to all physiological evaluation coefficients. The state prediction model is trained until the sum of prediction errors converges, at which point training stops. The state prediction model is a deep neural network model.

[0060] Furthermore, the comparison results include light sleep optimization instructions, deep sleep optimization instructions, dream optimization instructions, and wakefulness optimization instructions; the preset sleep duration includes a first sleep duration. Second sleep duration Third sleep duration and fourth sleep duration ;

[0061] Methods for obtaining the corresponding sleep duration based on the user's sleep state within the T+a time period include:

[0062] Record the sleep state output by the state prediction model;

[0063] If the sleep state is light sleep, then record the duration of the user's light sleep as follows: ,Will and To make a comparison, if If so, then no light sleep optimization instructions will be generated. Then, a light sleep optimization instruction will be generated;

[0064] If the sleep state is deep sleep, then record the user's deep sleep duration as follows: ,Will and To make a comparison, if If so, a deep sleep optimization instruction is generated. If so, no deep sleep optimization instructions will be generated;

[0065] If the sleep state is a dream state, then the duration of the user's dream is recorded as follows: ,Will and To make a comparison, if Then, a dream optimization command is generated. If not, dream optimization instructions will not be generated;

[0066] If the sleep state is a wake-up state, then record the user's wake-up duration as follows: ,Will and To make a comparison, if Then a wake-up optimization instruction is generated if If so, no wake-up optimization instruction will be generated.

[0067] Furthermore, methods for optimizing the initial sound playback data based on the comparison results to obtain sound control strategies include:

[0068] Step d1: Obtain the current user's sleep state, take the initial sound volume in the initial sound playback data as a fixed quantity, the initial playback duration as a variable, and mark it as the current adjustment value E;

[0069] Step d2: Let E = E + W, and record the physiological assessment coefficient at the current regulation value E, 0 < W < 3 hours;

[0070] Step d3: Repeat step d2. When the adjustment value E equals the preset playback duration threshold, obtain J physiological evaluation coefficients and jump to step d4, where J is an integer greater than zero.

[0071] Step d4: Take the initial playback duration in the initial sound playback data as a fixed quantity, the initial sound volume as a variable, and mark it as the current adjustment value F;

[0072] Step d5: Let F = F + M, and record the physiological assessment coefficient at the current regulation value F, where 0 < M < 70 dB;

[0073] Step d6: Repeat step d5. When the adjustment value F equals the preset sound volume threshold, obtain L physiological evaluation coefficients, where L is an integer greater than zero.

[0074] Step d7: Accumulate J physiological evaluation coefficients and L physiological evaluation coefficients to obtain R physiological evaluation coefficients, and sort the R physiological evaluation coefficients in ascending order of value;

[0075] Step d8: Use the volume adjustment value and playback duration adjustment value corresponding to the physiological evaluation coefficient with the smallest value as the sound control strategy.

[0076] Secondly, the present invention provides a sleep aid system that simulates dynamic natural sounds, comprising:

[0077] The data analysis module is used to acquire users' historical preference data and external environment data, and to generate a first reference coefficient based on the users' historical preference data and external environment data. The historical preference data includes sound type, volume value and playback duration.

[0078] The initial setup module is used to acquire the user's historical sleep cycle data, and input the first reference coefficient and the historical sleep cycle data into the pre-built first sound training model to obtain the user's initial sound playback data.

[0079] The status determination module is used to obtain the user's physiological assessment coefficient within the time period T, and determine the user's sleep status within the time period T+a based on the physiological assessment coefficient.

[0080] The optimization control module obtains the corresponding sleep duration based on the user's sleep state within the T+a time period, compares the corresponding sleep duration with the preset sleep duration, and optimizes the initial sound playback data based on the comparison results to obtain a sound control strategy.

[0081] The technical effects and advantages of this invention are as follows:

[0082] 1. This invention, through in-depth analysis of users' historical preference data, covering sound type, volume, and playback duration, can accurately match individual preferences, greatly improving user comfort and compliance. Simultaneously, by incorporating external environmental data, such as temperature and humidity differences and sound intensity, it fully considers the impact of environmental factors on sleep. The generated first reference coefficient provides a scientific quantitative basis for subsequent strategies, making sleep assistance more relevant to real-world situations. Further optimization of initial sound playback data using historical sleep cycle data allows for targeted adjustments based on the user's past sleep patterns, comprehensively achieving personalized sleep assistance customization to meet the diverse needs of different users.

[0083] 2. This invention determines sleep state based on real-time acquired physiological assessment coefficients and compares them with preset sleep durations to optimize sound playback strategies, achieving dynamic intelligent control for sleep assistance. During sleep, it can adjust sound parameters, such as volume and playback duration, according to the actual duration of the user's light sleep, deep sleep, dream state, or awake state. By continuously testing and sorting the physiological assessment coefficients under different adjustment values, the optimal sleep control strategy is determined, effectively improving sleep quality, helping users better enter a suitable sleep stage, reducing sleep disorders, and improving overall sleep health, thus creating an intelligent and scientific sleep improvement solution for users. Attached Figure Description

[0084] Figure 1 This is a flowchart of the sleep aid method using dynamic natural sound simulation in Example 1;

[0085] Figure 2 This is a flowchart of the method for obtaining historical sleep cycle data of users in Example 1;

[0086] Figure 3 This is a flowchart illustrating the sound control strategy in Example 1;

[0087] Figure 4 This is a schematic diagram of the sleep aid system for dynamic natural sound simulation in Example 2. Detailed Implementation

[0088] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0089] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0090] It should be understood that although terms such as "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and a similar second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.

[0091] Example 1

[0092] Please see Figure 1 As shown, this embodiment discloses a sleep aid method based on dynamic natural sound simulation, which is applied to a sleep aid device; the method includes:

[0093] Step 1: Obtain the user's historical preference data and external environment data, and generate a first reference coefficient based on the user's historical preference data and external environment data. The external environment data includes the temperature difference, humidity difference, and sound intensity of the user per unit time.

[0094] It should be understood that the sleep aid device is equipped with an environmental data sensor and a sound effect preference setting function. The environmental data sensor includes a temperature sensor, a humidity sensor, a fiber optic sensor, and a sound sensor. The temperature sensor is used to sense the ambient temperature and optimize comfort. The humidity sensor is used to monitor air humidity and maintain a suitable sleep environment. The fiber optic sensor is used to detect the room light intensity and adjust the sound effect or wake-up function. The sound sensor is used to sense ambient noise and dynamically adjust the sleep aid sound effect.

[0095] It should be noted that the historical preference data includes sound type, volume level, and playback duration; these are set by the user according to personal preferences using the sleep aid device and recorded in the database. The database also pre-stores natural sound effects such as the sound of a babbling brook, gentle wind, and soothing piano melodies.

[0096] In practice, methods for generating the first reference coefficient based on users' historical preference data and external environment data include:

[0097] Step s1: Analyze the historical preference data to obtain the historical preference coefficient. ;

[0098] Step s2: Analyze the external environmental data to obtain the environmental assessment coefficient. ;

[0099] Step s3: Let the current time be t, and the historical preference coefficient sequence be { }, The historical preference coefficient at the current time t;

[0100] The environmental assessment coefficient sequence is { }, The environmental assessment coefficient is given at the current time t; the attenuation factor is... , 1;

[0101] Step s4: Calculate the weighted sum of historical preference coefficients r = 1, 2, ..., t;

[0102] And the weighted sum of the environmental assessment coefficients. ;

[0103] Step s5: Calculate the historical preference coefficients. With environmental assessment coefficient After comprehensive processing, the first reference coefficient is obtained, and its calculation formula is as follows:

[0104]

[0105] In the formula, Indicates the first reference coefficient. and As a weighting factor, and .

[0106] The method for obtaining the historical preference coefficient includes:

[0107] Step a1: There are n different sound types, and each sound type corresponds to a technical score. , i = 1, 2, ..., n;

[0108] Step a2: Based on the user's selection of sound type frequency Perform weighted adjustments, the This represents the proportion of times the i-th sound type was selected out of the total number of selections.

[0109] Step a3: The volume value V is non-linearly mapped according to a logarithmic relationship. Let the minimum volume value be... The maximum volume value is Its mapping formula is:

[0110]

[0111] In the formula, This represents the volume value obtained after performing a non-linear mapping on the volume value V.

[0112] Step a4: The sound duration C is processed using a piecewise function, let... , , For different duration thresholds; when hour, ;when hour, ;when hour, ;when hour, , This refers to the playback duration.

[0113] Step a5: Select the sound type Selecting the frequency Volume value after non-linear mapping and playback duration Perform formulaic calculations to obtain historical preference coefficients;

[0114]

[0115] In the formula, This represents the historical preference coefficient.

[0116] The method for obtaining the environmental assessment coefficient includes:

[0117] Step b1: Temperature standard deviation The Gaussian function is used for processing, and the standard temperature range is set as []. , The real-time temperature is The calculation formula is:

[0118]

[0119] Step b1: Humidity standard deviation The Gaussian function is used for processing, and the standard humidity range is set as []. , The real-time humidity is The calculation formula is:

[0120]

[0121] Step b3: Standard deviation of sound intensity The process is performed using a power function, and the standard sound intensity range is set as []. , The real-time sound intensity is The calculation formula is:

[0122]

[0123] Step b4: Environmental Assessment Coefficient The calculation formula is:

[0124]

[0125] In the formula, Indicates the environmental assessment coefficient. , and As a weighting factor, and .

[0126] Different users have very different requirements for their sleep environment. This step takes into account external environmental data (such as indoor temperature, humidity, and sound intensity) and historical preference data to generate a first reference coefficient, which helps to adjust sleep aid strategies according to each user's specific situation.

[0127] Step 2: Obtain the user's historical sleep cycle data, and input the first reference coefficient and historical sleep cycle data into the pre-built first sound training model to obtain the user's initial sound playback data;

[0128] It should be noted that: the historical sleep cycle data includes each sleep state and its corresponding sleep duration, the sleep state includes light sleep state, deep sleep state, dream state and awake state; the initial sound playback data includes the initial music type, initial volume and initial playback duration.

[0129] It is worth noting that each sleep state includes a corresponding set of initial sound playback data. Specifically, the initial sound playback data for light sleep includes a first initial music type, a first initial volume, and a first initial playback duration; for deep sleep, it includes a second initial music type, a second initial volume, and a second initial playback duration; for dreaming, it includes a third initial music type, a third initial volume, and a third initial playback duration; and for awakening, it includes a fourth initial music type, a fourth initial volume, and a fourth initial playback duration.

[0130] Please see Figure 2 As shown, in practice, the methods for obtaining users' historical sleep cycle data include:

[0131] Step c1: Obtain historical EEG data, construct a time-domain EEG graph with time in the historical EEG data as the horizontal axis and EEG waves in the historical EEG data as the vertical axis. The EEG waves include theta waves, delta waves, beta waves and alpha waves.

[0132] It should be noted that the theta wave, delta wave, beta wave, and alpha wave of the brain are collected by the user during sleep using sleep monitoring equipment.

[0133] Step c2: Divide the EEG time-domain map according to the EEG to obtain a historical EEG set, which includes Z historical EEG waveforms, including theta wave waveforms, delta wave waveforms, beta wave waveforms and alpha wave waveforms.

[0134] Step c3: Extract the z-th historical EEG waveform from the historical EEG set, where z is a positive integer and the initial value of z is 1;

[0135] Step c4: Extract the corresponding standard EEG waveform, calculate the similarity between the historical EEG waveform and the standard EEG waveform. If the similarity between the historical EEG waveform and the standard EEG waveform is greater than or equal to the preset EEG similarity threshold, extract the corresponding sleep duration and jump to step c5; if the similarity between the historical EEG waveform and the standard EEG waveform is less than the preset EEG similarity threshold, jump directly to step c5.

[0136] It should be noted that: standard EEG waveforms include EEG waveforms for different sleep stages. For example, irregular theta waves in a standard EEG waveform indicate low-frequency brainwave activity, which indicates a light sleep state; delta waves (slow waves) in a standard EEG waveform indicate that the user's body is almost inactive and the heart rate is low, which indicates a deep sleep state; historical EEG waveforms showing relatively active beta or alpha waves, accompanied by rapid eye movements and irregular breathing, indicate a light sleep state with muscle paralysis, which indicates a dream state; historical EEG waveforms showing relatively chaotic and frequent patterns, accompanied by higher-frequency brainwaves (beta waves), indicate an arousal state.

[0137] Furthermore, the similarity algorithm includes cosine similarity algorithm or Euclidean distance algorithm, etc.; before calculating the similarity between historical EEG waveforms and standard EEG waveforms, preprocessing is required, including image enhancement, image denoising, and image segmentation, etc.

[0138] Step c5: Let z = z + 1, and jump back to step c3;

[0139] Step c6: Repeat steps c3 to c5 until z=Z, then end the loop and obtain the corresponding sleep duration.

[0140] In implementation, the method for constructing the first sound training model includes:

[0141] Historical sound training data is divided into a sound training set and a sound test set to construct a regression network model. The historical sound training data includes sound feature data and corresponding initial sound playback data. The sound feature data includes a first reference coefficient and historical sleep cycle data. The sound feature data in the sound training set is used as the input to the regression network model, and the initial sound playback data corresponding to the sound feature data in the sound training set is used as the output of the regression network model. The regression network model is trained to obtain an initial first regression network model. The training objective is to minimize the sum of prediction accuracies. The initial first regression network model is evaluated using the sound test set. If the sum of prediction accuracies is less than a preset accuracy threshold, the corresponding initial first regression network model is used as the first sound training model. If the sum of prediction accuracies is greater than or equal to the preset accuracy threshold, the initial first regression network model is trained again using the original sound training set until the test results meet the set threshold. The formula for calculating the sum of prediction accuracies is: In the formula: This represents the sum of prediction accuracies. Indicates the first sound test set Predicted values ​​of the initial sound playback data for the group. Indicates the first sound test set The actual value of the initial sound playback data for the group. Indicates the number of groups.

[0142] It should be noted that the initial first regression network model includes decision tree regression network model, random forest regression network model, support vector machine regression network model, linear regression network model or neural network model.

[0143] Step 3: Obtain the user's physiological assessment coefficient within time period T, and determine the user's sleep state within time period T+a based on the physiological assessment coefficient;

[0144] In practice, the methods for obtaining the user's physiological assessment coefficient within time period T include:

[0145] Acquire real-time physiological representation data of users, including the absolute value of heart rate difference, the absolute value of respiratory rate difference, brain wave type, and the absolute value of muscle electrical activity frequency difference;

[0146] It should be noted that the absolute values ​​of the heart rate difference, respiratory rate difference, and muscle electrical activity frequency difference are all calculated by comparing the real-time heart rate, respiratory rate, and muscle electrical activity frequency with the corresponding preset values. The smaller the absolute value of the difference, the more stable the user's physiological assessment coefficient and the better the sleep state.

[0147] Real-time physiological characterization data are normalized to obtain physiological assessment coefficients; the calculation formula is as follows:

[0148]

[0149] In the formula, Represents the physiological assessment coefficient. This represents the absolute value of the heart rate difference. This represents the absolute value of the difference in respiratory rates. Indicates the type of brainwaves. This represents the absolute value of the difference in muscle electrical activity frequencies. , , and As a weighting factor, and .

[0150] It should be noted that: before calculating the physiological assessment coefficient, the type of brainwave was considered. Preprocessing is performed, assuming , , and For different frequency band power thresholds; when hour, ;when hour, ;when hour, ;when hour, ; The frequency power of the delta wave (0.5-3Hz). The frequency power of the theta wave (4-7Hz). This represents the frequency power of the alpha wave (8-13Hz). This represents the frequency power of the β wave (14-30Hz).

[0151] It should be noted that the sleep aid device also connects to various physiological monitoring devices via wired or wireless means. These physiological monitoring devices include heart rate sensors, respiratory sensors, electroencephalography (EEG) devices, and electromyography (EMG) sensors. The specific models and types of these sensors are not limited. EEG devices detect the user's brain activity; brain waves mainly include alpha waves, beta waves, theta waves, and delta waves. Heart rate monitoring devices (such as smart bracelets or chest-worn heart rate monitors) record heart rate. During sleep, heart rate gradually decreases, remaining relatively stable and low during deep sleep, while slight fluctuations may occur during dreaming. Heart rate is relatively higher during wakefulness or light sleep. Respiratory monitoring devices (such as nasal clip-on respiratory sensors or pressure sensors under the mattress) are used to acquire respiratory rate. Generally, respiratory rate slows down during sleep, becoming more regular and slow during deep sleep. The stability of respiratory rate, such as the standard deviation of respiratory rate over a time interval T, can be calculated as a component of the physiological assessment coefficient. EMG sensors (usually attached near facial or limb muscles) monitor muscle activity. During deep sleep and some light sleep stages, muscle activity is minimal, while during REM sleep (dreaming), most muscles in the body, except for the eye muscles, are usually paralyzed.

[0152] In practice, methods for determining a user's sleep state within the T+a time period based on physiological assessment coefficients include:

[0153] Physiological assessment coefficients are input into a pre-built state prediction model to obtain the user's sleep state during the T+a time period.

[0154] Specifically, the training process of the state prediction model includes:

[0155] Physiological assessment coefficients of group P are collected in advance, where P is an integer greater than 1. Corresponding prediction results are set for each physiological assessment coefficient. These prediction results include light sleep, deep sleep, dream state, and arousal state. Different numerical labels are assigned to each of these states. For example, a numerical label of 1 is assigned to light sleep, 2 to deep sleep, 3 to dream state, and 4 to arousal state. The prediction results corresponding to the physiological assessment coefficients are collected by those skilled in the art during the historical physiological assessment coefficient diagnosis process. Based on practical experience, those skilled in the art sequentially determine the corresponding prediction results for each of the different physiological assessment coefficients in group P under different conditions.

[0156] The numerical labels of the prediction results are marked as prediction labels, and the physiological evaluation coefficients and their corresponding prediction labels are converted into a set of feature vectors.

[0157] Each set of feature vectors is used as input to the state prediction model, which outputs a set of predicted labels corresponding to each set of physiological evaluation coefficients and uses the actual predicted labels corresponding to each set of physiological evaluation coefficients as the prediction target. The actual predicted labels are the pre-set predicted labels corresponding to the physiological evaluation coefficients. The training objective is to minimize the sum of prediction errors of the predicted labels corresponding to all physiological evaluation coefficients. The state prediction model is trained until the sum of prediction errors converges, at which point training stops. The state prediction model is specifically a deep neural network model.

[0158] Step 4: Based on the user's sleep state during the T+a time period, obtain the corresponding sleep duration, compare the corresponding sleep duration with the preset sleep duration, optimize the initial sound playback data based on the comparison results, and obtain the sound control strategy.

[0159] It should be noted that the comparison results include light sleep optimization instructions, deep sleep optimization instructions, dream optimization instructions, and wakefulness optimization instructions. The preset sleep duration includes the first sleep duration. Second sleep duration Third sleep duration and fourth sleep duration The preset sleep duration is set based on the experience of those skilled in the art, and the preset sleep duration is a number greater than zero.

[0160] In practice, methods for obtaining the corresponding sleep duration based on the user's sleep state within the T+a time period include:

[0161] Record the sleep state output by the state prediction model;

[0162] If the sleep state is light sleep, then record the duration of the user's light sleep as follows: ,Will and To make a comparison, if If so, then no light sleep optimization instructions will be generated. Then, a light sleep optimization instruction will be generated;

[0163] If the sleep state is deep sleep, then record the user's deep sleep duration as follows: ,Will and To make a comparison, if If so, a deep sleep optimization instruction is generated. If so, no deep sleep optimization instructions will be generated;

[0164] If the sleep state is a dream state, then the duration of the user's dream is recorded as follows: ,Will and To make a comparison, if Then, a dream optimization command is generated. If not, dream optimization instructions will not be generated;

[0165] If the sleep state is a wake-up state, then record the user's wake-up duration as follows: ,Will and To make a comparison, if Then a wake-up optimization instruction is generated if If so, no wake-up optimization instruction will be generated;

[0166] It should be noted that: settings The reason for generating the light sleep optimization instruction is that if the user exceeds the preset first sleep duration, it means that the user has been in a light sleep state. Being in a light sleep state for a long time will lead to insufficient secretion of growth hormone, which will affect the repair of various organs and tissues in the body, making it difficult to eliminate physical fatigue, which is not good for the user's physical and mental health.

[0167] Please see Figure 3 As shown, in practice, the method for optimizing the initial sound playback data based on the comparison results to obtain the sound control strategy includes:

[0168] Step d1: Obtain the current user's sleep state, take the initial sound volume in the initial sound playback data as a fixed quantity, the initial playback duration as a variable, and mark it as the current adjustment value E;

[0169] Step d2: Let E = E + W, and record the physiological assessment coefficient at the current regulation value E, 0 < W < 3 hours;

[0170] Step d3: Repeat step d2. When the adjustment value E equals the preset playback duration threshold, obtain J physiological evaluation coefficients and jump to step d4, where J is an integer greater than zero.

[0171] Step d4: Take the initial playback duration in the initial sound playback data as a fixed quantity, the initial sound volume as a variable, and mark it as the current adjustment value F;

[0172] Step d5: Let F = F + M, and record the physiological assessment coefficient at the current regulation value F, where 0 < M < 70 dB;

[0173] Step d6: Repeat step d5. When the adjustment value F equals the preset sound volume threshold, obtain L physiological evaluation coefficients, where L is an integer greater than zero.

[0174] Step d7: Accumulate J physiological evaluation coefficients and L physiological evaluation coefficients to obtain R physiological evaluation coefficients, and sort the R physiological evaluation coefficients in ascending order of value;

[0175] Step d8: Use the volume adjustment value and playback duration adjustment value corresponding to the physiological evaluation coefficient with the smallest value as the sound control strategy.

[0176] This embodiment, through in-depth analysis of users' historical preference data, covering sound type, volume, and playback duration, can accurately match individual preferences, greatly improving user comfort and compliance. Simultaneously, by incorporating external environmental data, such as temperature and humidity differences and sound intensity, it fully considers the impact of environmental factors on sleep. The generated first reference coefficient provides a scientific quantitative basis for subsequent strategies, making sleep assistance more relevant to real-world situations. Further optimization of initial sound playback data using historical sleep cycle data allows for targeted adjustments based on the user's past sleep patterns, achieving comprehensive personalized sleep assistance customization to meet the diverse needs of different users.

[0177] This embodiment determines sleep state based on real-time acquired physiological assessment coefficients and optimizes sound playback strategies by comparing them with preset sleep durations, achieving dynamic and intelligent control of sleep assistance. During sleep, it can adjust sound parameters, such as volume and playback duration, according to the actual duration of the user's light sleep, deep sleep, dream state, or awake state. By continuously testing and sorting the physiological assessment coefficients under different adjustment values, the optimal sleep control strategy is determined, effectively improving sleep quality, helping users better enter a suitable sleep stage, reducing sleep disorders, and improving overall sleep health, thus creating an intelligent and scientific sleep improvement solution for users.

[0178] Example 2

[0179] Please see Figure 4As shown, this embodiment provides a sleep aid system based on dynamic natural sound simulation. The system includes: a data analysis module, an initial setting module, a state determination module, and an optimization and control module. The modules are connected via wired and / or wireless means to achieve data transmission between the modules.

[0180] The data analysis module is used to acquire users' historical preference data and external environment data, and to generate a first reference coefficient based on the users' historical preference data and external environment data. The historical preference data includes sound type, volume value and playback duration.

[0181] The initial setup module is used to acquire the user's historical sleep cycle data, and input the first reference coefficient and the historical sleep cycle data into the pre-built first sound training model to obtain the user's initial sound playback data.

[0182] The status determination module is used to obtain the user's physiological assessment coefficient within the time period T, and determine the user's sleep status within the time period T+a based on the physiological assessment coefficient.

[0183] The optimization control module obtains the corresponding sleep duration based on the user's sleep state within the T+a time period, compares the corresponding sleep duration with the preset sleep duration, and optimizes the initial sound playback data based on the comparison results to obtain a sound control strategy.

[0184] Example 3

[0185] This embodiment provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the above-described embodiment of the sleep assistance method using dynamic natural sound simulation, for example... Figure 1 The flowchart is shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments, for example... Figure 4 The system structure diagram is shown below.

[0186] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0187] The memory is an internal storage unit of the terminal device, such as a hard drive or RAM. The memory can also be an external storage device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard. Furthermore, the memory may include both internal and external storage units. The memory is used to store the computer program and other programs and data required by the terminal device. The memory can also be used to temporarily store data that has been output or will be output.

[0188] Example 4

[0189] This embodiment provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the sleep aid method of dynamic natural sound simulation of Embodiment 1.

[0190] The formulas mentioned above are all dimensionless calculations, derived from software simulation using a large amount of data to approximate the real situation. The weighting factors and preset thresholds in the formulas are set by those skilled in the art based on the actual situation or obtained through large-scale data simulation. The size of the weighting factor is a specific value obtained by quantifying each parameter to facilitate subsequent comparison. The size of the weighting factor depends on the amount of sample data and the processing coefficients initially set by those skilled in the art for each set of sample data. As long as it does not affect the proportional relationship between the parameter and the quantified value, it is acceptable.

[0191] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0192] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A sleep aid method based on dynamic natural sound simulation, characterized in that, include: The system acquires the user's historical preference data and external environment data, and generates a first reference coefficient based on the user's historical preference data and external environment data. The historical preference data includes sound type, volume value, and playback duration. Acquire the user's historical sleep cycle data, and input the first reference coefficient and historical sleep cycle data into the pre-built first sound training model to obtain the user's initial sound playback data; Obtain the user's physiological assessment coefficient within time period T, and determine the user's sleep state within time period T+a based on the physiological assessment coefficient; Based on the user's sleep state during the T+a time period, the corresponding sleep duration is obtained, and the corresponding sleep duration is compared with the preset sleep duration. The initial sound playback data is optimized based on the comparison results to obtain a sound control strategy. External environmental data includes the temperature difference, humidity difference, and sound intensity of the user per unit time. Methods for generating a first reference coefficient based on user historical preference data and external environment data include: Step s1: Analyze the historical preference data to obtain the historical preference coefficient. ; Step s2: Analyze the external environmental data to obtain the environmental assessment coefficient. ; Step s3: Let the current time be t, and the historical preference coefficient sequence be { }, The historical preference coefficient at the current time t; The environmental assessment coefficient sequence is { }, The environmental assessment coefficient is given at the current time t; the attenuation factor is... , 1; Step s4: Calculate the weighted sum of historical preference coefficients r = 1, 2, ..., t; And the weighted sum of the environmental assessment coefficients. ; Step s5: Calculate the historical preference coefficients. With environmental assessment coefficient After comprehensive processing, the first reference coefficient is obtained, and its calculation formula is as follows: ； In the formula, Indicates the first reference coefficient. and As a weighting factor, and ; The methods for obtaining the historical preference coefficients include: Step a1: There are n different sound types, and each sound type corresponds to a technical score. , i = 1, 2, ..., n; Step a2: Based on the user's selection of sound type frequency Perform weighted adjustments, the This represents the proportion of times the i-th sound type was selected out of the total number of selections. Step a3: The volume value V is non-linearly mapped according to a logarithmic relationship. Let the minimum volume value be... The maximum volume value is Its mapping formula is: ； In the formula, This represents the volume value after a non-linear mapping of the volume value V. Step a4: The sound duration C is processed using a piecewise function, let... , , For different duration thresholds; when hour, ;when hour, ;when hour, ;when hour, , This refers to the playback duration. Step a5: Select the sound type Selecting the frequency Volume value after non-linear mapping and playback duration Perform formulaic calculations to obtain historical preference coefficients; ； In the formula, Represents the historical preference coefficient; Methods for obtaining environmental assessment coefficients include: Step b1: Temperature standard deviation The Gaussian function is used for processing, and the standard temperature range is set as []. , The real-time temperature is The calculation formula is: ； Step b1: Humidity standard deviation The Gaussian function is used for processing, and the standard humidity range is set as []. , The real-time humidity is The calculation formula is: ； Step b3: Standard deviation of sound intensity The process is performed using a power function, and the standard sound intensity range is set as []. , The real-time sound intensity is The calculation formula is: ； Step b4: Environmental Assessment Coefficient The calculation formula is: ； In the formula, Indicates the environmental assessment coefficient. , and As a weighting factor, and 1.

2. The sleep-aiding method based on dynamic natural sound simulation according to claim 1, characterized in that, The historical sleep cycle data includes each sleep state and its corresponding sleep duration, and the sleep states include light sleep, deep sleep, dream state and wakefulness state. The initial sound playback data includes the initial music type, initial volume, and initial playback duration; Methods for obtaining users' historical sleep cycle data include: Step c1: Obtain historical EEG data, construct a time-domain EEG graph with time in the historical EEG data as the horizontal axis and EEG waves in the historical EEG data as the vertical axis. The EEG waves include theta waves, delta waves, beta waves and alpha waves. Step c2: Divide the brainwave time-domain map according to the brainwaves to obtain a historical brainwave set, which includes Z historical brainwave waveforms, including theta wave waveforms, delta wave waveforms, beta wave waveforms and alpha wave waveforms; Step c3: Extract the z-th historical EEG waveform from the historical EEG set, where z is a positive integer and the initial value of z is 1; Step c4: Extract the corresponding standard EEG waveform, calculate the similarity between the historical EEG waveform and the standard EEG waveform. If the similarity between the historical EEG waveform and the standard EEG waveform is greater than or equal to the preset EEG similarity threshold, extract the corresponding sleep duration and jump to step c5; if the similarity between the historical EEG waveform and the standard EEG waveform is less than the preset EEG similarity threshold, jump directly to step c5. Step c5: Let z = z + 1, and jump back to step c3; Step c6: Repeat steps c3 to c5 until z=Z, then end the loop and obtain the corresponding sleep duration.

3. The sleep aid method based on dynamic natural sound simulation according to claim 2, characterized in that, The method for constructing the first sound training model includes: Historical sound training data is divided into a sound training set and a sound test set to construct a regression network model. The historical sound training data includes sound feature data and corresponding initial sound playback data. The sound feature data includes a first reference coefficient and historical sleep cycle data. The sound feature data in the sound training set is used as the input to the regression network model, and the initial sound playback data corresponding to the sound feature data in the sound training set is used as the output of the regression network model. The regression network model is trained to obtain an initial first regression network model with the training objective of minimizing the sum of prediction accuracies. The initial first regression network model is evaluated using the sound test set. If the sum of prediction accuracies is less than a preset accuracy threshold, the corresponding initial first regression network model is used as the first sound training model. If the sum of prediction accuracies is greater than or equal to the preset accuracy threshold, the initial first regression network model is trained again using the original sound training set until the test results meet the set threshold. The initial first regression network model can be a decision tree regression network model, a random forest regression network model, a support vector machine regression network model, a linear regression network model, or a neural network model.

4. The sleep-aiding method based on dynamic natural sound simulation according to claim 3, characterized in that, Methods for obtaining the user's physiological assessment coefficient within time period T include: Acquire real-time physiological representation data of users, including the absolute value of heart rate difference, the absolute value of respiratory rate difference, brain wave type, and the absolute value of muscle electrical activity frequency difference; Real-time physiological characterization data are normalized to obtain physiological assessment coefficients; Methods for determining a user's sleep state within the T+a time period based on physiological assessment coefficients include: The physiological assessment coefficients are input into a pre-built state prediction model to obtain the user's sleep state during the T+a time period. The specific training process of the state prediction model includes: P sets of physiological evaluation coefficients are collected in advance, where P is an integer greater than 1. The corresponding prediction results are set for the physiological evaluation coefficients. The prediction results include light sleep state, deep sleep state, dream state and awake state. Different numerical labels are set for light sleep state, deep sleep state, dream state and awake state. The numerical labels of the prediction results are marked as prediction labels. The physiological evaluation coefficients and the corresponding prediction labels are converted into a set of feature vectors. Each set of feature vectors is used as input to the state prediction model. The state prediction model outputs a set of predicted labels corresponding to each set of physiological evaluation coefficients and uses the actual predicted labels corresponding to each set of physiological evaluation coefficients as the prediction target. The actual predicted labels are the pre-set predicted labels corresponding to the physiological evaluation coefficients. The training objective is to minimize the sum of prediction errors of the predicted labels corresponding to all physiological evaluation coefficients. The state prediction model is trained until the sum of prediction errors converges, at which point training stops. The state prediction model is a deep neural network model.

5. The sleep aid method based on dynamic natural sound simulation according to claim 4, characterized in that, The comparison results include light sleep optimization instructions, deep sleep optimization instructions, dream optimization instructions, and wakefulness optimization instructions; the preset sleep duration includes the first sleep duration. Second sleep duration Third sleep duration and fourth sleep duration ; Methods for obtaining the corresponding sleep duration based on the user's sleep state within the T+a time period include: Record the sleep state output by the state prediction model; If the sleep state is light sleep, then record the duration of the user's light sleep as follows: ,Will and To make a comparison, if If so, then no light sleep optimization instructions will be generated. Then, a light sleep optimization instruction will be generated; If the sleep state is deep sleep, then record the user's deep sleep duration as follows: ,Will and To make a comparison, if If so, a deep sleep optimization instruction is generated. If so, no deep sleep optimization instructions will be generated; If the sleep state is a dream state, then the duration of the user's dream is recorded as follows: ,Will and To make a comparison, if Then, a dream optimization command is generated. If not, dream optimization instructions will not be generated; If the sleep state is a wake-up state, then record the user's wake-up duration as follows: ,Will and To make a comparison, if Then a wake-up optimization instruction is generated if If so, no wake-up optimization instruction will be generated.

6. The sleep aid method based on dynamic natural sound simulation according to claim 5, characterized in that, The methods for optimizing the initial sound playback data based on the comparison results to obtain sound control strategies include: Step d1: Obtain the current user's sleep state, take the initial sound volume in the initial sound playback data as a fixed quantity, the initial playback duration as a variable, and mark it as the current adjustment value E; Step d2: Let E = E + W, and record the physiological assessment coefficient at the current regulation value E, 0 < W < 3 hours; Step d3: Repeat step d2. When the adjustment value E equals the preset playback duration threshold, obtain J physiological evaluation coefficients and jump to step d4, where J is an integer greater than zero. Step d4: Take the initial playback duration in the initial sound playback data as a fixed quantity, the initial sound volume as a variable, and mark it as the current adjustment value F; Step d5: Let F = F + M, and record the physiological assessment coefficient at the current regulation value F, where 0 < M < 70 dB; Step d6: Repeat step d5. When the adjustment value F equals the preset sound volume threshold, obtain L physiological evaluation coefficients, where L is an integer greater than zero. Step d7: Accumulate J physiological evaluation coefficients and L physiological evaluation coefficients to obtain R physiological evaluation coefficients, and sort the R physiological evaluation coefficients in ascending order of value; Step d8: Use the volume adjustment value and playback duration adjustment value corresponding to the physiological evaluation coefficient with the smallest value as the sound control strategy.

7. A sleep aid system based on dynamic natural sound simulation, used to implement the sleep aid method based on dynamic natural sound simulation as described in any one of claims 1-6, characterized in that, include: The data analysis module is used to acquire users' historical preference data and external environment data, and to generate a first reference coefficient based on the users' historical preference data and external environment data. The historical preference data includes sound type, volume value and playback duration. The initial setup module is used to acquire the user's historical sleep cycle data, and input the first reference coefficient and the historical sleep cycle data into the pre-built first sound training model to obtain the user's initial sound playback data. The status determination module is used to obtain the user's physiological assessment coefficient within the time period T, and determine the user's sleep status within the time period T+a based on the physiological assessment coefficient. The optimization control module obtains the corresponding sleep duration based on the user's sleep state within the T+a time period, compares the corresponding sleep duration with the preset sleep duration, and optimizes the initial sound playback data based on the comparison results to obtain a sound control strategy.

Images