Alarm clock function interaction control method and system for smart watch

Abstract

The invention relates to the field of intelligent watch interaction control, in particular to an alarm clock function interaction control method and system for an intelligent watch. The alarm clock function interaction control method for the smart watch comprises the following steps: S10, setting forward shaking time, and when acquiring that a user is in a waking state from an acquired wrist physiological signal, acquiring wrist movement information as reference information within the forward shaking time before an alarm clock is triggered; according to the scheme, the closing intention is recognized by collecting the wrist motion information, extracting the numerical value and frequency change features and comparing the feature differences before and after the alarm clock is triggered, the limitation of a fixed gesture input instruction is broken through, the closing instruction input requirements of different gestures in multiple scenes are met, and the user experience is improved. The problem that the user experience is low due to the fact that the mode of closing the alarm clock of the intelligent watch through a touch screen does not cover the use requirement of the user sufficiently is effectively solved, and finally natural, accurate and personalized alarm clock closing interaction is achieved.

Description

Technical Field

[0001] This invention relates to the field of smartwatch interactive control, and in particular to a method and system for interactive control of the alarm clock function in a smartwatch. Background Technology

[0002] With the widespread adoption of smart wearable devices, smartwatches, thanks to their portability and multi-functional integration, have become a core tool for users' daily time management. The alarm clock function, as a fundamental and frequently used application, covers diverse needs, including daily wake-up calls to help users smoothly transition from sleep to wakefulness, and also fragmented time reminders such as meeting duration management at work, Pomodoro timers for studying, and cooking time monitoring, helping users efficiently plan their time and improve task processing efficiency. Therefore, the practicality and ease of use of the alarm clock function directly impact the overall user experience of a smartwatch.

[0003] Currently, alarm clock control on smartwatches is mostly based on touchscreen interaction. For example, the gesture-controlled alarm clock method and smart terminal disclosed in Chinese patent application CN104102444A manages the alarm clock by detecting touchscreen signals. For instance, when the alarm is triggered, if a touch signal with a long press duration ≥ a preset threshold (e.g., 2 seconds) is detected, the alarm is turned off; if a touch signal with a swipe distance ≥ a preset threshold (e.g., 2 cm) is detected, the nap function is activated. This method does not require users to accurately identify the alarm clock menu on the screen; it can be operated simply by long press or swipe. This is particularly suitable for users who are groggy and have imprecise physical movements in the morning, effectively avoiding the inconvenience of traditional precise clicks (on the smartwatch screen) and improving the ease of use of the morning alarm clock.

[0004] However, users' needs for smartwatch alarms are not limited to morning alarms; daily timer alarms are also frequently used, and the core needs of the two differ significantly. When using timer alarms, users mostly only need to be reminded of the end of a specific task at a preset time (such as the end of a meeting, the coffee being brewed, or a cooking milestone), without needing the sustained and powerful wake-up notification of a morning alarm. More importantly, when a timer alarm is triggered, users are often in a focused state (such as using a computer for work, operating kitchen utensils for cooking, or holding a book for reading), and their hands may be occupied or unable to leave the object being operated. The success rate of touch control on capacitive screens drops significantly when hands are wet, gloves are worn, or tools are held, making it difficult or inconvenient to turn off the alarm by touching the screen. This results in existing touch control methods not adequately covering users' needs. This inadequate coverage of needs causes the alarm to ring continuously, interfering with the user's current task and significantly reducing the user experience in timer alarm scenarios. Summary of the Invention

[0005] This invention provides an interactive control method and system for the alarm clock function of a smartwatch, in order to solve the problem that the method of turning off the alarm clock on a smartwatch by touch screen does not adequately cover the user's needs, resulting in a poor user experience.

[0006] To solve the above-mentioned technical problems, this application provides the following technical solution: The interactive control method for the alarm clock function of a smartwatch includes the following steps: S10: Set the pre-animation time. When the user is awake as determined from the collected wrist physiological signals, wrist movement information is acquired during the pre-animation time before the alarm is triggered as reference information. Wrist movement information after the alarm is triggered is acquired in real time as comparison information. The change characteristics of the reference information and comparison information are analyzed in combination with the acquisition time. The change characteristics include numerical change characteristics and frequency change characteristics. S20: Determine whether the change characteristics of the reference information conform to the preset instruction characteristics. If: S20a: If not met, the alarm will be turned off when the change characteristics of the comparison information match the preset instruction characteristics; S20b: If it meets the requirements, extract the change features of the reference information as the reference feature, extract the change features of the comparison information as the comparison feature, and comprehensively evaluate the differences between the reference feature and the comparison feature based on the numerical change features and frequency change features of the comparison feature and the reference feature. When the comprehensive evaluation result meets the preset instruction features, turn off the alarm clock.

[0007] The basic principle and beneficial effects of this solution are as follows: When the user is awake, this solution collects and analyzes wrist movement information (such as the amplitude of acceleration fluctuations collected by a smartwatch) and frequency change characteristics (such as the dynamic pattern of vibration cycles) before and after the alarm is triggered. Then, by comparing the differences in wrist movement information before and after the alarm is triggered, it determines whether the user intends to turn off the alarm. In this process, this solution does not limit operations to specific gestures (such as fixed finger taps or swipes), but rather captures the user's natural actions through the "comparison of changes in movement characteristics (numerical and frequency characteristics) obtained from the hand (wrist). Regardless of whether the user expresses the intention to turn off the alarm through tapping, shaking, or adjusting grip strength, as long as the numerical and frequency change characteristics of the wrist movement information before and after the alarm are triggered conform to the "active operation for the alarm" (i.e., the preset instruction characteristics) pattern (i.e., the user's behavioral attitude towards turning off the alarm after it is triggered is different from their current focus), it can be identified. For example, when a user is taking notes in a meeting (in a conscious state), the smartwatch recognizes the regular changes (including numerical and frequency changes) formed by holding the pen during the pre-call period. After the alarm is triggered, the user, not wanting to put down the pen while working on the meeting notes, changes the rhythm of the movement by quickly and slightly shaking their wrist (e.g., the frequency changes from 1Hz for steady writing to 3Hz for rapid shaking, and the numerical characteristics change from stable to a sudden increase). The smartwatch compares the difference between the reference characteristics and the comparison characteristics, recognizes this as a deliberate response to the alarm, and turns off the alarm. This design breaks through the limitations of traditional touch or fixed gestures, better conforming to the user's natural behavioral habits in different scenarios (such as small movements when both hands are occupied), making the acquisition of the turn-off command more accurate and greatly improving the user experience without deliberate cooperation. Furthermore, it can also specifically distinguish between daily habitual actions and alarm-related responses, enhancing adaptability to complex scenarios.

[0008] Secondly, this solution analyzes numerical change characteristics (such as the increasing or decreasing trend of signal strength) and frequency change characteristics (such as the abrupt change pattern of vibration frequency), and combines this with the vibration when the alarm is triggered. This allows for more accurate capture of the user's operational intentions under the interference of alarm vibration. For example, when a user's hands are occupied while cooking (e.g., holding kitchen utensils), the hands have already been in continuous motion due to stirring before the alarm is triggered (including specific numerical and frequency change characteristics). When the alarm vibrates, the user issues a shutdown command by adjusting their grip (e.g., squeezing or releasing the kitchen utensils). The vibration amplifies the numerical fluctuations and frequency abrupt changes caused by the grip change, enabling the smartwatch to accurately distinguish between normal stirring actions and deliberate grip adjustments by recognizing these amplified change characteristics. This avoids misinterpreting everyday actions as shutdown commands. Furthermore, it eliminates the need for the user to interrupt cooking to touch the screen or perform specific gestures, reducing the difficulty of cooperating when recording the alarm shutdown command. This further solves the problem of inconvenient touch control in focused scenarios and improves the user experience.

[0009] Finally, because the solution extracts personalized numerical and frequency variation features (such as the reference features of the pre-wake time), it can naturally adapt to the operating habits of different users, solving the problem of recognition failure caused by individual movement differences. For example, elderly users may be accustomed to using a large wrist swing to turn off the alarm (numerical variation features are high amplitude fluctuations, frequency variation features are low frequency), while younger users may prefer rapid fingertip taps (numerical variation features are high frequency small amplitude fluctuations, frequency variation features are high frequency). The smartwatch evaluates the differences by comparing the reference features of the pre-wake time with the characteristics after triggering, and can accurately recognize the intention without requiring users to learn a uniform standard action. This is difficult to achieve with traditional fixed gesture control, further improving the user experience for different users.

[0010] In summary, this solution identifies the intention to turn off the alarm by collecting wrist movement information, extracting numerical and frequency change features, and comparing the feature differences before and after the alarm is triggered. This not only breaks through the limitations of fixed gesture input commands but also adapts to the needs of inputting different gestures for turning off the alarm in various scenarios. It effectively solves the problem that the method of turning off the smartwatch alarm by touch screen does not cover the user's needs and results in a poor user experience. Ultimately, it achieves a natural, accurate, and personalized alarm-turning interaction.

[0011] Further, in step S20b, the periodic characteristics of the changes in the reference information and the comparison information are analyzed. Based on the periodic characteristics, some wrist movement information is extracted from the tail of the reference information as tail similarity information, and some wrist movement information is extracted from the head of the comparison information as head similarity information. The differences between the changes in the head similarity information and the tail similarity information are comprehensively evaluated. If the comprehensive evaluation result is less than a preset comparison threshold, some wrist movement information is extracted from the tail of the comparison information as comparison similarity information based on the periodic characteristics of the comparison features. The differences between the changes in the head similarity information and the comparison similarity information are comprehensively evaluated. When the comprehensive evaluation result is greater than the preset comparison threshold, the alarm clock is turned off.

[0012] This solution analyzes the periodic characteristics of reference and comparison information, extracts similar information in stages, and evaluates differences. This allows for precise differentiation between the continuation of a user's focused task and the closing gesture, improving recognition accuracy. Before the alarm is triggered, the user handles tasks manually (such as assembling parts). After triggering, the focused state continues, and the difference in change characteristics between the tail of the reference information and the head of the comparison information is small. This solution first compares the similarity of the change characteristics between the tail of the reference information and the head of the comparison information to obtain the user's task continuation status. Then, based on the difference between the head and tail of the comparison information, it accurately identifies the moment the user transitions to the closing gesture and promptly turns off the alarm. Simultaneously, this solution can filter transitional actions from task continuation to recording the closing gesture (such as minor adjustments to the action rhythm), avoiding misjudgment of transitional actions while accurately capturing the user's intention to turn off the alarm. This improves the adaptability of the solution to various scenarios, enhancing recognition stability and user experience.

[0013] Furthermore, in step S20b, the comprehensive evaluation result is correlated with the acquisition time of the comparison information, the rate of change of the comprehensive evaluation result over time is analyzed, and the time length for comparing similar information is adjusted according to the rate of change.

[0014] This solution can shorten the time for collecting similar information under high-speed change conditions, reduce the invalid duration of alarm ringing, improve interactive response efficiency, and avoid interfering with the user's need to quickly process the shutdown command. At the same time, it retains the original time for low-speed change scenarios, which can completely capture the action features of the user during the hesitation process, avoid misjudgment caused by incomplete collection, and does not require additional sensor hardware costs. It can achieve the dual effect of efficiency improvement and recognition accuracy through time correlation analysis alone, adapting to the user's usage needs under different action transition rhythms.

[0015] Furthermore, when the rate of change of the comprehensive evaluation result over time is less than a preset speed threshold, the rate of change of the change characteristics of the comparative similar information over time is obtained. If the rate of change is still less than the preset speed threshold, the difference in change characteristics between the comparative similar information and the reference information is taken as a suspected instruction feature. When the suspected instruction feature matches the preset instruction feature, the alarm clock is turned off.

[0016] Before the alarm is triggered, if the wrist movement data value or frequency is high, and the user cannot provide sufficient coordination to turn off the alarm (such as in a scenario where the user is stirring with a spatula), it is difficult to obtain the alarm-off command simply by analyzing the differences in wrist movement signals after the alarm is triggered, given the large hand movement data. In this case, by comparing the change features of similar information (the wrist movement information most likely for the user to input the gesture to turn off the alarm), and eliminating the change features of reference information (the user's wrist movement information before the alarm is triggered), we can obtain the wrist movement information when the user inputs the gesture to turn off the alarm, thus more accurately obtaining the user's actual intention. In this way, this solution does not require additional acquisition and calculation of other parameters, reducing computing power consumption and achieving the triple effects of accurate recognition, scenario adaptation, and cost control, ensuring that users can efficiently turn off the alarm without interrupting cooking.

[0017] Furthermore, when the value of the suspected instruction characteristic is small, the vibration intensity of the alarm clock is gradually increased.

[0018] When the user is focused on a particular scenario (such as holding kitchen utensils while cooking or using a computer while working) and the value of the suspected instruction is small, the intensity of the alarm clock vibration is gradually increased. This can amplify the motion characteristic signal of the weak shutdown action, solve the problem of difficulty in recognizing instructions when both hands are occupied, and avoid the initial strong vibration suddenly interfering with the focused state (such as not interrupting the rhythm while reading), which fits the timing needs that do not require strong wake-up.

[0019] Furthermore, in step S10, with the permission obtained, sound information is collected, semantically related sound information is extracted from the sound information as speech information, the proportion of speech information in the sound information is analyzed, and the vibration intensity when the alarm is triggered is adjusted according to the proportion.

[0020] When a user is focused (such as operating kitchen utensils for cooking or using a computer for work as mentioned in the background technology), the alarm clock's vibration intensity is adjusted by collecting and extracting sound information and analyzing its proportion. On the one hand, in scenarios where the proportion of sound information is high (such as when a user is communicating with others in a meeting or listening to audio explanations while cooking), the vibration intensity is automatically reduced to avoid strong vibrations interfering with the user's current voice interaction or information acquisition, thus meeting the need for a powerful wake-up alarm. On the other hand, in scenarios where the proportion of sound information is low (such as when a user is focused on reading without voice input or working in a quiet environment), the vibration intensity is appropriately increased to ensure that the user can perceive the alarm reminder, solving the problem in the background technology where alarm reminders are easily ignored when touch control is inconvenient. Meanwhile, this solution does not rely on user hand operation. It can dynamically adapt to the reminder needs of different environments by analyzing sound information alone. It avoids the limitations of touch control in scenarios where both hands are occupied. It also achieves the dual effect of interference control and reminder effectiveness through a single sound acquisition and analysis function. Especially in scenarios where it is inconvenient for users to touch the screen and the ambient sound status is changing, it can automatically match the vibration intensity and improve the user experience in multiple scenarios. This is something that traditional alarm clock control methods with fixed vibration intensity or reliance on touch control cannot achieve.

[0021] Furthermore, in step S10, if no physiological electrical signal is collected during the current rocking time, sound information is collected during the pre-rocking time as the pre-rocking sound. The frequency change characteristics and amplitude change characteristics of the pre-rocking sound are extracted, and the numerical change characteristics and periodic change characteristics of the reference information are also extracted. The correlation between the pre-rocking sound and the change characteristics of the reference information is analyzed through a feature matching algorithm, and the tightness of the correlation is calculated. The influence values ​​of the sound information and the reference information on turning off the alarm are set according to the tightness. After the alarm is triggered, the change characteristics of the sound information before and after the alarm is triggered are used as the sound change characteristics. In step S20, after comprehensive evaluation, the comprehensive evaluation results are adjusted according to the relationship between the influence values, the sound change characteristics, and the comprehensive evaluation results.

[0022] In scenarios where physiological electrical signals cannot be collected (i.e., when the smartwatch is not worn), the system collects the pre-shaking sound, extracts its features, analyzes the correlation with reference information features, and sets an influence value. After the alarm is triggered, the system adjusts the comprehensive evaluation result based on the sound change characteristics. This solves the problem of inconvenient touch control and improves the compatibility of command recognition through multi-signal collaboration. For example, when a user is cooking, holding kitchen utensils makes it impossible to collect physiological electrical signals. The steady-state sound of the range hood in the pre-shaking sound is correlated with the reference information of the wrist tossing motion. The smartwatch sets a reasonable influence value based on this. After the alarm is triggered, if the sound change (such as the range hood pausing) and the difference between the motion characteristics meet the evaluation criteria, the alarm can be turned off without touch control.

[0023] Furthermore, in step S10, after collecting the pre-rocking sound, the pre-rocking sound is divided into different types based on the collection time, including steady-state type and transient type, and the timbre features corresponding to different types of sound are extracted; in step S20b, if the comprehensive evaluation result does not meet the preset instruction features, the preset vibration mode of the alarm clock is gradually adjusted, and the comprehensive evaluation result is recalculated after adjusting the vibration mode; when the comprehensive evaluation result meets the preset instruction features, the sound type and timbre features collected this time are associated and stored with the adjusted vibration mode to form user habit data; a deep learning model is constructed based on the user habit data, taking the sound type and timbre features as input, and outputting the appropriate vibration mode, thereby setting the vibration mode when the alarm clock is triggered.

[0024] This solution not only addresses the difficulty users face in controlling the alarm clock after it's triggered in different scenarios, but also allows the smartwatch to autonomously adapt to user habits. It's particularly adaptable across industry scenarios. For example, in the catering industry, when chefs are working, the deep learning model can output an appropriate vibration pattern based on the steady-state sound of the range hood and the transient sound of the spatula hitting the pot. This avoids the problem of traditional fixed vibrations being easily masked by ambient noise or interfering with cooking actions, and eliminates the need for repeated manual adjustments by the user. It ensures the accuracy of the alarm clock interaction, enhances the smartwatch's versatility across different industry scenarios, and gradually aligns the vibration pattern with user preferences through the accumulation of habit data, achieving both personalized and scenario-based adaptation.

[0025] Furthermore, in step S10, when extracting the timbre features of steady-state sound, the harmonic distribution is determined simultaneously by combining its frequency change features; when extracting transient sound, the peak impact energy is determined by combining the amplitude change features; the correlation between sound features and reference information is strengthened or weakened by harmonic distribution and peak impact energy.

[0026] This scheme combines the steady-state sound harmonic distribution determined by frequency change characteristics to quantify its matching pattern with the frequency change characteristics of the reference information (wrist movement), avoiding miscorrelation caused by a single timbre feature. Combined with the transient sound impact energy peak determined by amplitude change characteristics, it can accurately anchor the temporal synchronization with the numerical change characteristics of the reference information, reducing environmental noise interference and making the calculation of the correlation tightness more reliable. Simultaneously, by strengthening effective correlations (such as the matching of stable harmonics with regular motion) and weakening ineffective correlations (such as the misalignment of chaotic harmonics with random motion), the sound timbre features are adaptively adjusted according to the actual correlation strength, avoiding weight allocation bias and providing more reasonable basic parameters for the comprehensive evaluation in step S20b. Attached Figure Description

[0027] Figure 1 This is a flowchart of the interactive control method for the alarm clock function of a smartwatch in Embodiment 1 of this solution. Detailed Implementation

[0028] The following will describe the concept and technical effects of the present invention clearly and completely with reference to embodiments, so as to fully understand the purpose, features and effects of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the scope of protection of the present invention. Example 1 like Figure 1 As shown, the interactive control method for the alarm clock function of a smartwatch includes the following steps: S10: Set the pre-animation time. When the user is awake based on the collected wrist physiological signals, acquire wrist movement information as reference information during the pre-animation time before the alarm is triggered, and collect wrist movement information after the alarm is triggered in real time as comparison information. Combine the acquisition time to analyze the change characteristics of the reference information and the comparison information; the change characteristics include numerical change characteristics and frequency change characteristics.

[0029] S20: Determine whether the change characteristics of the reference information conform to the preset instruction characteristics (determined by the administrator based on calculation accuracy, including preset numerical instruction characteristics and frequency instruction characteristics, where numerical instruction characteristics include signal amplitude standard deviation exceeding the threshold per unit time, slope absolute value reaching the set value, etc.; frequency instruction characteristics include main frequency offset exceeding the standard, high frequency energy ratio reaching the specified proportion, etc.; the preset instruction characteristics are used to identify whether the user has hand use from the wrist physiological signals collected before the alarm clock is triggered through numerical change characteristics and frequency change characteristics), if: S20a: If not, the alarm will be turned off when the change characteristics of the comparison information match the preset instruction characteristics. When the change characteristics of the comparison information do not match the preset instruction characteristics, the user's hand usage did not change significantly before and after the alarm was triggered. In this case, the user may not have moved their hands before and after the alarm was triggered, or the hand movements may not have changed significantly. Numerical instruction characteristics also include time characteristics. If the user may not have moved their hands before and after the alarm was triggered, then the change characteristics of the comparison information match the preset instruction characteristics; if the user moved their hands before and after the alarm was triggered (the user can only collect information on wrist movement by using a watch; if the alarm is triggered at this time, the user can feel the wrist vibration or hear the alarm ring), after the user knows that the alarm has been triggered, the duration of the user's hand movements matching the frequency instruction characteristics (time characteristics) can be used to determine whether the user has actively entered the instruction to turn off the alarm.

[0030] S20b: If the condition is met (i.e., it contains information about the user's hand usage), then extract the change features of the reference information (the change features of the user's hand usage before the alarm is triggered) as the reference features, and extract the change features of the comparison information (the change features of the user's hand usage after the alarm is triggered) as the comparison features. Based on the numerical change features and frequency change features of the comparison features and the reference features (comparing the changes in the user's hand usage before and after the alarm is triggered), a comprehensive evaluation is performed on the differences between the reference features and the comparison features. When the comprehensive evaluation result meets the preset instruction features, the alarm is turned off.

[0031] Smartwatches integrate gyroscopes and accelerometers into their hardware. They primarily use optical sensors to emit light that penetrates the skin, receiving the reflected light and converting it into electrical signals. By comparing these signals between sleep and wakefulness, the watch detects differences: during sleep, the heart rate is slower and more stable, resulting in a flatter waveform; while during wakefulness, the sympathetic nervous system is more active, leading to a slightly higher heart rate, greater fluctuations, and a steeper waveform. Furthermore, analysis of heart rate variability (HRV) reveals a higher proportion of high-frequency components during wakefulness; signal stability is also assessed, as subtle wrist movements introduce specific noise during wakefulness; and rest periods are considered to aid in the assessment. Some smartwatches also incorporate ECG sensors, using electrocardiogram waveform characteristics to enhance accuracy and ultimately precisely determine the user's wakefulness, providing a basis for collecting motion information during pre-wake cycles.

[0032] Among them, the numerical change characteristics focus on the dynamic fluctuations of the time domain signal. That is, the original signal is first segmented by sliding window method, denoised by low-pass filtering and normalized, and then the fluctuation intensity (standard deviation, root mean square), peak features (number of peaks and difference), trend features (signal slope) and other indicators are extracted. For example, when a user holds a spatula to stir-fry, the standard deviation is stable, but when the grip force is adjusted, the standard deviation rises sharply. This is used to quantify the change in the strength and amplitude of the wrist movement.

[0033] Frequency variation features focus on mining periodic patterns in the frequency domain. Specifically, this involves: first, performing a Fourier Transform (FFT) on the preprocessed time-domain signal to convert it into a frequency domain representation of frequency and energy (e.g., sampling rate of 50-200Hz to avoid spectral aliasing); then extracting features such as the dominant frequency (including the peak frequency with the highest energy percentage, such as 1Hz when writing and 3Hz when shaking the wrist), the frequency band energy ratio (e.g., the energy ratio between low and high frequency ranges), and frequency stability (e.g., spectral entropy). For example, during deliberate manipulation, the high-frequency energy percentage increases from 20% to over 50%, thus capturing the difference in wrist movement rhythm and cycle. Combining these two types of features (the combination of numerical and frequency features) can accurately distinguish between the movement differences of continuous activity and deliberate manipulation, providing a quantitative basis for alarm clock off command recognition.

[0034] In step S20b, the periodic characteristics of the changes in the reference information and the comparison information are analyzed. Based on the periodic characteristics, some wrist movement information is extracted from the tail of the reference information as tail similarity information, and some wrist movement information is extracted from the head of the comparison information as head similarity information. The differences between the changes in the head similarity information and the tail similarity information are comprehensively evaluated. If the comprehensive evaluation result is less than the preset comparison threshold (set by the administrator), some wrist movement information is extracted from the tail of the comparison information as comparison similarity information based on the periodic characteristics of the comparison features. The differences between the changes in the head similarity information and the comparison similarity information are comprehensively evaluated. When the comprehensive evaluation result is greater than the preset comparison threshold, the alarm clock is turned off.

[0035] When conducting a comprehensive evaluation, the differences in characteristics are first quantified, specifically including the differences in numerical variation characteristics, frequency variation characteristics, and the difference in the proportion of high-frequency energy.

[0036] When quantifying differences in numerical variation characteristics, the influence of individual movement amplitude differences is eliminated by calculating the normalized difference rate between the reference and comparison characteristics. (Numerical characteristic normalized difference rate) ,in, The mean of the numerical characteristics of the reference information (such as the standard deviation of amplitude, peak value difference, slope, etc.); To compare the mean of numerical features of the information. In the above process, by calculating the ratio of the absolute value of the difference to the mean of the numerical features of the reference information, the differences between different users and different action amplitudes are standardized to ensure horizontal comparability and intuitively reflect the significant changes in numerical features.

[0037] When quantifying the differences in frequency variation characteristics, further quantification of the changes in motion cycle and energy distribution is achieved by focusing on calculating the difference between the main frequency offset and the proportion of high-frequency energy. (Main frequency offset) ,in The primary frequency (the frequency with the highest energy percentage, measured in Hz) is used as a reference. The primary frequency for comparison information. Frequency offset directly reflects changes in movement rhythm, enabling smartwatches to clearly capture core differences in frequency characteristics.

[0038] High-frequency energy ratio difference ;in The energy percentage of high-frequency bands (e.g., 5-10Hz) is used as a reference. To compare the proportion of high-frequency energy in information. The difference in the proportion of high-frequency energy can reflect the change in the distribution of action energy. The proportion of high-frequency energy for deliberately turning off actions is usually higher. This value can be used to help distinguish between "transaction continuation" (i.e., the user continues the operation before the alarm is triggered after the alarm is triggered) and "turn-off command" (i.e. the gesture made by the user to turn off the alarm).

[0039] After quantifying the differences in features, the quantified differences are combined with scenario adaptability to adjust feature weights and generate a comprehensive evaluation value. , The calculation formula is shown in formula (1) below. (1), in The numerical feature weights are set by the administrator (for example, a base value of 0.4, adjusted to 0.5 for static, focused scenes). The frequency feature weight is set by the administrator (for example, the base value is 0.6, and it is adjusted to 0.7 in the vibration interference scenario). The unnormalized comprehensive evaluation value is (normalized value ranges from 0 to 1). The normalized comprehensive evaluation value S is... ,in, This is the minimum reference value preset for administrators (based on a large number of user sample statistics, for example, fixed at 0.2, corresponding to the typical original value in the "transaction continuation" scenario). The maximum reference value preset for the administrator (e.g., fixed at 1.8, corresponding to the typical original value in the "deliberately turned off" scenario). Normalization of the comprehensive evaluation value can eliminate the influence of differences in units and numerical ranges of the original features, ensuring that the comprehensive evaluation values ​​of different scenarios and different users are comparable.

[0040] Formula (1) highlights the contribution of key features through dynamic weighting. For example, during vibration interference, frequency features exhibit stronger anti-interference capabilities, thus improving... To a minimum of 0.7, thus avoiding the impact of environmental noise on the assessment results.

[0041] Finally, personalized thresholds are calibrated based on fluctuations in users' daily behavior to reduce misjudgments. Personalized thresholds ,in K is the baseline threshold (e.g., a fixed value of 0.6); K is the calibration coefficient (e.g., a fixed value of 0.1). The standard deviation of the reference information features (reflecting the degree of fluctuation in a user's daily actions) is used to personalize the calculation and adapt to individual differences in actions, such as when a user's daily actions fluctuate greatly. Set to 0.3 to avoid misjudgment due to unstable movement amplitude; when When the alarm is turned off, it is determined to be a valid shutdown command.

[0042] In the above, the preset comparison threshold is a fixed judgment standard set in advance based on a large number of user samples (covering different scenarios and action types). The personalized threshold is based on the preset comparison threshold (e.g., Based on 0.6, and combined with the individual user's action fluctuation characteristics (such as the standard deviation of daily actions), The threshold is dynamically calculated and may vary for each user.

[0043] In step S20b, the comprehensive evaluation results are correlated with the acquisition time of the comparison information, the rate of change of the comprehensive evaluation results over time is analyzed, and the time length for comparing similar information is adjusted according to the rate of change.

[0044] Specifically, the system first synchronously records the timestamps of the comprehensive evaluation results and corresponding comparison information, forming data that correlates evaluation results with time. Then, it calculates the ratio of the difference between evaluation results at consecutive time points to the time interval to obtain the rate of change (unit: normalized evaluation value / second). If the rate of change is greater than or equal to a preset threshold (e.g., 0.2, specifically set by the administrator), it indicates that the user's intention to switch from transaction continuation to closing is clear, and the comparison time for similar information is shortened (e.g., from 5 seconds to 2-3 seconds) to reduce invalid alarm rings. If the rate of change is less than the threshold, the original duration (e.g., 5 seconds) is retained to ensure complete feature collection, avoid misjudgment, and balance response efficiency and recognition accuracy.

[0045] In practice, User A uses a smartwatch to set a timer for writing. User A first sets a 30-second pre-writing time. During the writing process, the smartwatch collects physiological and motion signals from the user's wrist. The smartwatch collects wrist physiological signals through an optical heart rate sensor. Assuming the smartwatch detects a heart rate of 72 beats / min (within the normal range for a conscious state) and a high-frequency component of heart rate variability of 35% (beyond the sleep threshold), it determines that the user is awake. Then, 30 seconds before the alarm is triggered, the watch uses a six-axis sensor (accelerometer + gyroscope) to collect wrist movement information as a reference, filtering out noise and extracting numerical variation features (including amplitude standard deviation and peak value difference) and frequency variation features (including dominant frequency and high-frequency energy percentage). After the alarm is triggered, the smartwatch collects wrist movement information in real time for comparison. For example, the user continues writing for the first 5 seconds, and then adjusts their movements for the next 5 seconds due to the alarm ringing.

[0046] Next, the smartwatch determines whether the reference information features match the preset instruction features (where numerical features include amplitude standard deviation and peak value difference; frequency features include main frequency and high-frequency energy percentage). Assuming the reference information features match the preset instruction features, the S20b branch is executed. The smartwatch extracts the reference features and comparison features (e.g., motion features within 15 seconds after the alarm is triggered, and numerical and frequency features in the following 5 seconds), comprehensively evaluates the differences between the two, calculates a comprehensive evaluation value based on the numerical and frequency change features, and normalizes it, assuming the result is 0.925. Combining this with the user's historical action data (assuming a standard deviation of 0.2 for daily writing actions), the calculated personalized threshold is assumed to be 0.612. Because the comprehensive evaluation value of 0.925 is greater than the personalized threshold of 0.612, it is determined that the preset instruction features are met, and the alarm is triggered to turn off.

[0047] Example 2 The only difference between this embodiment and embodiment 1 is that when the rate of change of the comprehensive evaluation result over time is less than the preset speed threshold (set by the administrator according to the individual circumstances of the user; the reaction speed of young people is generally higher than that of the elderly, so the preset speed threshold for young people is generally higher than that for the elderly), the rate of change of the change characteristics of the comparative similar information over time is obtained. If the rate of change is still less than the preset speed threshold, the difference in the change characteristics between the comparative similar information and the reference information is taken as the suspected instruction characteristics. When the suspected instruction characteristics meet the preset instruction characteristics, the alarm clock is turned off.

[0048] When the value of the suspected instruction characteristic is small, gradually increase the vibration intensity of the alarm clock.

[0049] Specifically, when the value of the suspected instruction feature is less than the preset feature value threshold (the preset feature value threshold is set based on the fluctuation range of the reference information feature and the preset lower limit of the instruction feature), the vibration intensity of the alarm clock is gradually increased according to the preset gradient. After each adjustment of the vibration intensity, the change features of the comparison similar information are reacquired and the suspected instruction feature is updated. If the updated suspected instruction feature matches the preset instruction feature, the increase of vibration intensity is stopped and the alarm clock is turned off. If the suspected instruction feature still does not match the preset instruction feature after adjusting the vibration intensity multiple times (usually 3 times, depending on the user setting), the current vibration intensity is maintained and the collection time of the comparison similar information is extended.

[0050] In practice, when User 2 is cooking in the kitchen, they set an alarm for a soup simmering. At this time, the user needs to continuously toss the pan, making it impossible to make any obvious gestures; they can only try to turn off the alarm by tapping the pan handle.

[0051] Before the alarm is triggered, the smartwatch collects wrist movement information from the user while tossing the pan as a reference. The numerical and frequency changes of this information show strong regularity, consistent with the user's current high-frequency hand usage. After the alarm is triggered, the overall evaluation result shows that the rate of change over time is less than a preset speed threshold. Further comparison is made with the rate of change of similar information (i.e., motion information including the action of clicking the pan handle). This rate of change is still less than the threshold, so the difference between the similar information and the reference information is used as the suspected instruction feature.

[0052] Because the value of the suspected command feature was small, the smartwatch gradually increased the vibration intensity until the suspected command feature matched the preset command feature, successfully turning off the alarm. Throughout the process, the user did not need to interrupt the tossing operation, ensuring that the cooking process was not affected.

[0053] Example 3 The only difference between this embodiment and embodiments 1-2 is that in step S10, sound information is collected with permission, semantically related sound information is extracted from the sound information as speech information, the proportion of speech information in the sound information is analyzed, and the vibration intensity when the alarm clock is triggered is adjusted according to the proportion.

[0054] Specifically, after obtaining permission to collect sound information, the system first removes non-semantic interference noise from the environment (such as wind noise and equipment operation noise) through a noise filtering module, and then extracts semantically relevant sound information as speech information. When analyzing the proportion of speech information in the filtered sound information, if the proportion is not less than a preset proportion threshold (set by the user), it is determined that the user is in a voice interaction scenario (such as meeting communication or listening to audio tutorials), and the vibration intensity when the alarm is triggered is reduced to a range that does not interfere with voice reception. If the proportion is less than the preset proportion threshold, it is determined that the user is in a low-voice scenario (such as a noisy environment), and the vibration intensity when the alarm is triggered is increased to ensure user perception. Moreover, after each adjustment of the vibration intensity, the matching data of the speech information proportion and vibration intensity in the current scenario is recorded simultaneously, and the matching data can be directly called for quick adjustment in the same scenario in the future.

[0055] In practice, User C uses a smartwatch's timer alarm to remind participants of meeting intervals and document submission deadlines. Before a particular online project meeting, User C sets a meeting preparation reminder alarm to trigger one hour later; at this point, the smartwatch has already obtained permission to collect audio information.

[0056] During the meeting, the smartwatch uses a noise filtering module to remove non-semantic noises such as air conditioning operation and keyboard typing in real time, extracting user C's voice communication with colleagues and meeting presentation audio as voice information. Assuming that the analysis finds that the proportion of voice information in the filtered sound information is not less than a preset proportion threshold, the smartwatch intelligently determines that user C is in a voice interaction scenario, and then automatically adjusts the vibration intensity of the alarm clock to a low-intensity mode (e.g., only a slight sensation on the wrist, without producing a noticeable vibration sound) to avoid interfering with meeting communication when triggered.

[0057] After the meeting, User C entered document writing mode. With no significant voice interaction in the environment, the smartwatch analyzed the environment and determined the voice information ratio to be below a preset threshold, classifying it as a low-voice scenario. It then adjusted the alarm's vibration intensity to medium to ensure the reminder was still perceived. After each adjustment, the watch synchronously records the matching data between the voice ratio and vibration intensity for the current scenario. Subsequent times when User C is in a similar online meeting or document writing scenario, the watch can directly access historical matching data to quickly adjust the vibration intensity without needing to re-analyze, balancing reminder effectiveness with control over scene interference.

[0058] Example 4 The only difference between this embodiment and embodiments 1-3 is that, in step S10, when no physiological electrical signal is collected during the pre-shaking time, sound information is collected during the pre-shaking time as the pre-shaking sound, and the frequency change features and amplitude change features of the pre-shaking sound are extracted. At the same time, the numerical change features and periodic change features of the reference information are extracted. The correlation between the pre-shaking sound and the reference information change features is analyzed by feature matching algorithm, the tightness of the correlation is calculated, and the influence values ​​of the sound information and the reference information on turning off the alarm are set according to the tightness.

[0059] After the alarm is triggered, the changes in sound information before and after the alarm is triggered are taken as sound change features. In step S20, after comprehensive evaluation, the comprehensive evaluation results are adjusted according to the relationship between the influence value, sound change features and comprehensive evaluation results.

[0060] Specifically, when no physiological electrical signals are collected during the pre-shaking period, the smartwatch simultaneously collects sound information and wrist movement information. More specifically, ambient sound is first collected as the pre-shaking sound, and after filtering out interference, the frequency (fluctuation period, peak timing) and amplitude (rise and fall rate, stable duration) variation characteristics are extracted; the synchronously collected wrist movement is used as reference information, and the reference information and the pre-shaking sound value (amplitude fluctuation, frequency of sudden changes) and period (repetition duration, peak distribution) variation characteristics are extracted.

[0061] The dynamic time warping algorithm is used to align the feature curves of the reference information and the pre-rocking sound. The correlation is determined by frequency overlap and amplitude trend consistency. Then, the tightness is calculated by weighting the three dimensions of frequency coordination (curve overlap rate), amplitude coordination (correlation coefficient), and timing synchronization (peak time difference), and the influence value is set according to the tightness.

[0062] The pre-rock sound feature sequence is set as follows (m is the number of sampling points, (It is the eigenvalue of point i), and the reference information (wrist movement) sequence is: (n is the number of motion sampling points). The difference between individual feature points is calculated using Euclidean distance: Then construct an m×n distance matrix D. , represents the Euclidean distance between the feature of the i-th sampling point of the pre-rocking sound and the feature of the j-th sampling point of the reference information. The smaller the value, the higher the similarity.

[0063] Then, a dynamic programming algorithm is used to find the distance matrix. arrive Optimal path (k is the path length, m is the pre-rocking sound feature sequence) , where n is the feature sequence of reference information (e.g., wrist movement). ,satisfy ). Here It is the t-th point on the path, corresponding to a position in the distance matrix. ,So It is the element value of that position in the distance matrix D, that is, the Euclidean distance value of the t-th point on the path.

[0064] Use dynamic programming to find the path with the minimum cumulative distance. Where k is the path length. The DTW value represents the path point distance; the smaller the DTW value, the higher the alignment between the reference information and the pre-rock sound characteristic curve. Frequency overlap is used... calculate, It represents the frequency overlap, where k is the frequency characteristic sequence of the pre-rocking sound after alignment. and reference information frequency associated feature sequence The number of sampling points (because the two sequences have the same length after dynamic time warping and alignment, k represents the number of common sampling points). t is the index variable for summation, from 1 to k, which sequentially judges and sums the condition of each sampling point. It is an indicator function, when At the time of its establishment, The value is 1; when the condition is not met, the value is 0, which is used to count the number of sampling points that meet the frequency difference within the tolerance range. It is the frequency characteristic value of the pre-rock sound at the t-th sampling point, that is, the frequency value at the t-th position in the frequency sequence of the pre-rock sound after alignment. It is the frequency correlation feature value of the reference information at the t-th sampling point, that is, the frequency correlation value at the t-th position in the frequency correlation sequence of the reference information after alignment. It is a preset frequency tolerance threshold, used to define the range within which the frequency difference between the pre-rocking sound and the reference information can be considered "overlapping". For example, it can be set to a specific value such as 50Hz according to the actual needs of the scenario.

[0065] The amplitude trend is expressed using the Pearson coefficient. calculate. represents the Pearson correlation coefficient, used to quantify the degree of linear correlation between the pre-rock sound amplitude feature sequence and the reference information amplitude correlation feature sequence. k is the pre-rock sound amplitude feature sequence. and reference information amplitude associated feature sequence The number of sampling points (the two sequences have the same length after processing, denoted by k, representing the number of common sampling points). t is the index variable for summation, ranging from 1 to k, which calculates and sums the relevant data for each sampling point sequentially. It is the amplitude characteristic value of the pre-rock sound at the t-th sampling point, that is, the amplitude value at the t-th position in the pre-rock sound amplitude sequence. It is the amplitude correlation feature value of the reference information at the t-th sampling point, that is, the amplitude correlation value at the t-th position in the amplitude correlation sequence of the reference information. It is the mean of the pre-rock sound amplitude characteristic sequence, expressed by the formula The calculated value represents the average level of the pre-rock sound amplitude. It is the mean of the characteristic sequence associated with the reference information amplitude, obtained through the formula The calculated value represents the average level of the correlation with the reference information amplitude.

[0066] Total score for tightness , , , These are the weights for frequency coordination, amplitude coordination, and timing synchronization, with the sum of the weights being 1 (which can be fine-tuned according to actual scenario requirements to highlight the importance of different dimensions). The magnitude coherence is represented by the Pearson correlation coefficient. To obtain the non-negative value, that is... This eliminates the interference of negative correlation on amplitude coherence and focuses only on the consistency of amplitude trends of positive correlation. Indicates temporal synchronization degree, used to quantify the degree of synchronization between the pre-rocking sound and reference information in the time series. Its calculation method is typically as follows: ,in The characteristic peak mean time difference, This represents the maximum permissible time difference.

[0067] For different Different association levels can be set for the value. For example, S≥0.8 represents a very strong association level, which is the sound influence value. Reference impact value When 0.5 ≤ S < 0.8, the correlation is considered to be of medium level. , When S < 0.5, the association is considered weak. , ,and .

[0068] After the alarm is triggered, the changes in sound frequency and amplitude, as well as the persistence of features, are compared before and after the alarm. Following step S20b, the difference between the reference and real-time motion features is calculated to obtain an initial comprehensive evaluation result. Next, the matching degree between the sound change features and the preset command features is multiplied by a correction term for the sound influence value. The initial result is then multiplied by a motion term for the motion influence value. The correction term and the motion term are added together to obtain the adjusted result. If the result meets the standard, the alarm is turned off; otherwise, the influence value can be fine-tuned and recalculated based on sound stability to ensure recognition accuracy.

[0069] In step S10, after collecting the pre-rocking sound, the pre-rocking sound is divided into different types based on the collection time, including steady-state type and transient type, and the timbre features corresponding to different types of sound are extracted. In step S20b, if the comprehensive evaluation result does not meet the preset instruction features, the preset vibration mode of the alarm clock is gradually adjusted (set by the user, for example, adjusting the vibration mode of the alarm clock to a gradually increasing intensity mode), and the comprehensive evaluation result is recalculated after adjusting the vibration mode. When the comprehensive evaluation result meets the preset instruction features, the sound type and timbre features collected this time are associated and stored with the adjusted vibration mode to form user usage habit data. A deep learning model is built based on the user usage habit data, taking the sound type and timbre features as input and outputting the appropriate vibration mode, thereby setting the vibration mode when the alarm clock is triggered.

[0070] In step S10, when extracting the timbre features of steady-state sound, the harmonic distribution is determined simultaneously by combining its frequency change features; when extracting transient sound, the peak impact energy is determined by combining the amplitude change features; the correlation between sound features and reference information is strengthened or weakened by harmonic distribution and peak impact energy.

[0071] The harmonic frequency sequence of steady-state sound is as follows: ;in Let be the harmonic frequency of the t-th sampling point, extracted by combining the frequency variation characteristics of steady-state sound. The frequency sequence corresponding to the periodic variation characteristics of the reference information is: ,in The periodic correlation frequency of the t-th sampling point is obtained by converting the periodic variation characteristics of the reference information. The correlation formula between the two is: ,in This represents the correlation between the harmonic distribution and the periodic characteristics of the reference information, with a value range of [0,1]. The closer the value is to 1, the stronger the correlation. k is the number of sampling points of the two sequences after alignment (consistent with the "length of the aligned feature sequence"). t is the summation index variable, which iterates through each sampling point. This is the frequency sensitivity coefficient, used to adjust the sensitivity to frequency differences (e.g., it can be increased when there is a lot of ambient noise). (to enhance the identification of subtle differences) As an exponential function, it rapidly increases the correlation between similar frequencies through nonlinear mapping, while significantly reducing the correlation between frequencies with large differences, which aligns with the real-world principle that "closer frequencies have stronger correlations".

[0072] Let the peak energy sequence of transient sound be... ,in The peak impact energy at the t-th sampling point is extracted by combining the amplitude variation characteristics of the transient sound. The numerical variation feature sequence of the reference information is as follows. ,in Let be the numerical feature value of the t-th sampling point, directly obtained from the numerical change characteristics of the reference information. The correlation formula between the two is: ,in This represents the correlation between the peak impact energy and the numerical characteristics of the reference information, with a value range of [0,1]. The closer the value is to 1, the stronger the correlation. k is the number of sampling points in the two sequences after alignment. t is the summation index variable. The reference information is the time corresponding to the peak value of the numerical characteristics (the benchmark for judging the temporal synchronization with the peak value of the transient sound). This is the timing decay coefficient, used to control the impact of timing deviations on correlation (e.g., in fast-motion scenarios). Take the largest value to strictly limit timing deviations). This is a minimum value (e.g., 1e-6) used to avoid calculation errors where the denominator is 0; the first term Used to quantify timing synchronization; the greater the timing deviation, the smaller this value; the second term... Used to quantify amplitude coherence; the closer the amplitudes are, the closer this value is to 1.

[0073] The overall correlation between harmonic distribution, peak impact energy, and reference information is obtained by weighted fusion of the two dimensions mentioned above: ,in To determine the overall correlation, the value range is [0,1], reflecting the overall correlation strength between sound characteristics (harmonic distribution, peak impact energy) and reference information; As a weighting coefficient, it dynamically adapts to the sound type—when the proportion of steady-state sound is high. Taking a large value (e.g., 0.7) emphasizes the correlation contribution of harmonic distribution; when transient sound accounts for a high proportion... Take a smaller value (e.g., 0.3) to emphasize the correlation contribution of the impact energy peak; , These are the harmonic distribution correlation degree and the impact energy peak correlation degree mentioned above, respectively, to ensure that the parameters are consistent with the previous formula.

[0074] In practice, User D is an office worker who uses a smartwatch alarm clock to wake up. The default pre-arrest time for the smartwatch is 5 minutes before the alarm is triggered. On a weekday morning, User D was sleeping on his side when his wrist was under pressure. During the pre-arrest time, the watch did not collect physiological electrical signals. At this time, the watch automatically activated the sound and motion acquisition mode: it simultaneously collected steady-state sounds (air conditioner running sound) and transient sounds (the sound of the sheets rubbing when User D turned over) in the room as the pre-arrest sound, and collected wrist movements (turning over, raising the hand) as reference information. After data collection, the watch filters out interference from the pre-shaking sound and extracts the frequency variation characteristics (assuming a fluctuation period of 20Hz and stable peak timing) and amplitude variation characteristics (assuming a stable duration of 4 minutes and a gentle lifting and lowering rate) of the air conditioner sound. It also extracts the frequency variation characteristics (peak timing concentrated at the moment of turning over) and amplitude variation characteristics (assuming a fast lifting and lowering rate and a stable duration of 0.5 seconds) of the bed sheet friction sound. Simultaneously, it extracts the numerical variation characteristics (assuming amplitude fluctuation of ±30° and abrupt change frequency of 2 times) and periodic variation characteristics (assuming the lifting motion is repeated for 1.5 seconds and the peak is distributed at the moment of friction sound).

[0075] Next, the feature curves are aligned using the Dynamic Time Warping (DTW) algorithm: Let the sound feature sequence X (assuming m = 200 sampling points) and the motion feature sequence Y (assuming n = 180 sampling points) be used. A 200×180 distance matrix is ​​constructed using the Euclidean distance formula, and then the optimal path is found through dynamic programming. Assume the calculated DTW value is 0.12 (high alignment). Then, the frequency overlap is calculated, assuming... The amplitude Pearson coefficient is 0.92 (assuming the frequency difference between the air conditioner noise and the turning cycle is ≤50Hz). The value is 0.88 (the trend of friction noise is consistent with the amplitude of raising the hand), combined with the timing synchronization. It is 0.9, weighted. , , The density S is calculated with values ​​of 0.3, 0.3, and 0.4 respectively. Assuming S is 0.9, let... 0.5 It is 0.5.

[0076] After the alarm is triggered, user D claps (the sound changes with a sudden increase in frequency and a doubling of amplitude). Assuming the watch's initial evaluation result is 0.7, and then multiplied by the clapping accuracy matching the preset command (0.9),... Assuming a correction term of 0.45, the initial result is multiplied by... Assuming the exercise score is 0.35, the adjusted score is assumed to be 0.8 (meeting the target), and the alarm is turned off. If user D fails to meet the target score in a subsequent clapping session, the watch gradually adjusts the vibration to increase in intensity. Once the target score is met, the watch stores the transient clapping sound, corresponding timbre, and increasing vibration as habit data. After accumulating 10 data points, a deep learning model is built. When user D claps subsequently, the deep learning model directly outputs increasing vibration without repeated adjustments.

[0077] The above are merely embodiments of the present invention. The invention is not limited to the fields covered by these embodiments. Commonly known structures and characteristics in the solutions are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are able to access all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. An interactive control method for the alarm clock function of a smartwatch, characterized in that, Includes the following steps: S10: Set the pre-animation time. When the user is awake as determined from the collected wrist physiological signals, wrist movement information is acquired during the pre-animation time before the alarm is triggered as reference information. Wrist movement information after the alarm is triggered is acquired in real time as comparison information. The change characteristics of the reference information and comparison information are analyzed in combination with the acquisition time. The change characteristics include numerical change characteristics and frequency change characteristics. S20: Determine whether the change characteristics of the reference information conform to the preset instruction characteristics. If: S20a: If not met, the alarm will be turned off when the change characteristics of the comparison information match the preset instruction characteristics; S20b: If it meets the requirements, extract the change features of the reference information as the reference feature, extract the change features of the comparison information as the comparison feature, and comprehensively evaluate the differences between the reference feature and the comparison feature based on the numerical change features and frequency change features of the comparison feature and the reference feature. When the comprehensive evaluation result meets the preset instruction features, turn off the alarm clock.

2. The interactive control method for the alarm clock function of a smartwatch according to claim 1, characterized in that: In step S20b, the periodic characteristics of the change features of the reference information and the comparison information are analyzed. Based on the periodic characteristics, some wrist movement information is extracted from the tail of the reference information as tail similarity information, and some wrist movement information is extracted from the head of the comparison information as head similarity information. The difference between the change features of the head similarity information and the tail similarity information is comprehensively evaluated. If the comprehensive evaluation result is less than the preset comparison threshold, some wrist movement information is extracted from the tail of the comparison information as comparison similarity information based on the periodic characteristics of the comparison features. The alarm clock is turned off when the difference between the changes in head-similar information and the comparison-similar information is comprehensively evaluated and the comprehensive evaluation result is greater than the preset comparison threshold.

3. The interactive control method for the alarm clock function of a smartwatch according to claim 2, characterized in that: In step S20b, the comprehensive evaluation results are correlated with the acquisition time of the comparison information, the rate of change of the comprehensive evaluation results over time is analyzed, and the time length for comparing similar information is adjusted according to the rate of change.

4. The interactive control method for the alarm clock function of a smartwatch according to claim 3, characterized in that: When the rate of change of the comprehensive evaluation result over time is less than the preset rate threshold, the rate of change of the change characteristics of the comparative similar information over time is obtained. If the rate of change is still less than the preset rate threshold, the difference in the change characteristics between the comparative similar information and the reference information is taken as the suspected instruction characteristics. When the suspected instruction characteristics meet the preset instruction characteristics, the alarm clock is turned off.

5. The interactive control method for the alarm clock function of a smartwatch according to claim 4, characterized in that: When the value of the suspected instruction characteristic is small, gradually increase the vibration intensity of the alarm clock.

6. The interactive control method for the alarm clock function of a smartwatch according to claim 1 or 5, characterized in that: In step S10, with the permission obtained, sound information is collected, semantically related sound information is extracted from the sound information as speech information, the proportion of speech information in the sound information is analyzed, and the vibration intensity when the alarm clock is triggered is adjusted according to the proportion.

7. The interactive control method for the alarm clock function of a smartwatch according to claim 6, characterized in that: In step S10, if no physiological electrical signal is collected during the current rocking time, sound information is collected during the pre-rocking time as the pre-rocking sound, and the frequency change characteristics and amplitude change characteristics of the pre-rocking sound are extracted. At the same time, the numerical change characteristics and periodic change characteristics of the reference information are extracted. The correlation between the pre-earthquake sound and the change features of the reference information is analyzed by feature matching algorithm, the tightness of the correlation is calculated, and the influence values ​​of the sound information and the reference information on turning off the alarm are set according to the tightness. After the alarm is triggered, the changes in sound information before and after the alarm is triggered are used as sound change features. In step S20b, after conducting a comprehensive evaluation, the comprehensive evaluation results are adjusted based on the relationship between the impact value, sound change characteristics, and the comprehensive evaluation results.

8. The interactive control method for the alarm clock function of a smartwatch according to claim 7, characterized in that: In step S10, after the pre-rocking sound is collected, the pre-rocking sound is divided into different types based on the collection time, including steady-state type and transient type, and the timbre features corresponding to different types of sound are extracted at the same time. In step S20b, if the comprehensive evaluation result does not meet the preset instruction characteristics, the preset vibration mode of the alarm clock is gradually adjusted, and the comprehensive evaluation result is recalculated after adjusting the vibration mode. When the comprehensive evaluation result meets the preset instruction characteristics, the sound type and timbre characteristics collected this time are associated and stored with the adjusted vibration mode to form user habit data. A deep learning model is built based on the user habit data, taking the sound type and timbre characteristics as input and outputting the appropriate vibration mode, thereby setting the vibration mode when the alarm clock is triggered.

9. The interactive control method for the alarm clock function of a smartwatch according to claim 8, characterized in that: In step S10, when extracting the timbre features of steady-state sound, the harmonic distribution is determined simultaneously by combining its frequency change features; when extracting transient sound, the peak impact energy is determined by combining the amplitude change features; the correlation between sound features and reference information is strengthened or weakened by harmonic distribution and peak impact energy.

10. An interactive control system for the alarm clock function of a smartwatch, characterized in that, The alarm clock function interactive control method for smartwatches as described in any one of claims 1-9 is used.

Images