Relieving lamp intelligent control method and system based on voice interaction

Abstract

The invention discloses a soothing lamp intelligent control method and system based on voice interaction, and relates to the technical field of data encryption, and the method comprises the steps: receiving a voice command inputted by a user through a voice interaction device, awakening a control system, and starting a voice emotion recognition unit; the voice emotion recognition unit is used for performing emotion analysis on the voice command, extracting emotion characteristics and judging real-time emotion of the user; on the basis of continuous voice input of a user, voice signals are collected regularly, the emotional state is updated dynamically, and light setting is optimized in combination with environmental data; by recording historical voice interaction data and environment data of a user, an emotion mode and light preference of the user are learned, and light setting is continuously optimized. Light setting is continuously optimized through a self-learning mechanism, it is ensured that the system can conduct self-adaptive adjustment along with changes of user requirements, and light control is more intelligent and personalized.

Description

Technical Field

[0001] The present invention relates to the field of data encryption technology, and in particular to a method and system for intelligently controlling a soothing lamp based on voice interaction. Background Art

[0002] With the rapid development of smart home systems, voice interaction technology has gradually become one of the main ways of human-computer interaction and is widely used in smart devices such as lighting and temperature control, greatly improving the user experience. In the field of smart lighting control, traditional voice command-based control systems usually only support simple operations, such as turning lights on and off, adjusting brightness, and switching color temperature. Although these systems can achieve basic lighting management, their functions are often relatively simple and cannot dynamically respond to changes in the user's emotional state or the surrounding environment. Especially in scenarios where the user's emotions change, traditional lighting control systems cannot perceive the user's emotional needs through voice interaction, thereby providing a more personalized and context-aware lighting control solution.

[0003] The main problem with existing intelligent lighting control technology is that it lacks the ability to recognize voice emotions and cannot adjust lighting settings according to the user's real-time emotions. Although some systems have introduced environmental perception functions that can detect external conditions such as light and temperature, these functions have not been effectively combined with the user's emotional state, resulting in a relatively mechanical lighting adjustment process that lacks a humanized and intelligent experience. This deficiency makes the performance of intelligent lighting control systems very limited in improving user emotions and relieving emotional stress, making it difficult to meet users' higher-level needs. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a soothing lamp intelligent control method and system based on voice interaction to solve the problem in the prior art that the lighting settings cannot be dynamically adjusted based on the user's emotional state.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a soothing lamp intelligent control method based on voice interaction, which includes receiving a voice command input by a user through a voice interaction device, waking up a control system and starting a voice emotion recognition unit;

[0008] The speech emotion recognition unit performs emotion analysis on the speech commands, extracts emotion features and determines the user's real-time emotions;

[0009] Based on the determined emotional state, the environmental perception function is enabled to detect the external conditions of the surrounding environment in real time and generate corresponding lighting control parameters;

[0010] The generated lighting control parameters are transmitted to the lighting control device to adjust the brightness, color temperature and color settings of the soothing light in real time;

[0011] Based on the user's continuous voice input, voice signals are collected regularly, emotional status is dynamically updated, and lighting settings are optimized in combination with environmental data;

[0012] By recording the user's historical voice interaction data and environmental data, we learn the user's emotional patterns and lighting preferences and continuously optimize the lighting settings.

[0013] As a preferred solution of the soothing lamp intelligent control method based on voice interaction of the present invention, wherein: the receiving of the voice command input by the user through the voice interaction device, waking up the control system and starting the voice emotion recognition unit, the specific steps are:

[0014] The microphone array is used to beamform the voice signal, enhance the direction of the user's voice, and obtain a voice signal with a high signal-to-noise ratio.

[0015] Use Wiener filter combined with deep learning model to perform adaptive noise reduction on speech signals;

[0016] Perform short-time Fourier transform on the preprocessed speech signal to generate an STFT spectrum of the speech;

[0017] The CNN model is used to analyze the STFT spectrum graph to detect whether it contains the wake-up word. When the wake-up word is detected successfully, the system switches to the wake-up state and starts the emotion recognition unit.

[0018] As a preferred solution of the soothing light intelligent control method based on voice interaction described in the present invention, wherein: the voice emotion recognition unit performs emotion analysis on the voice command, extracts the emotion features and judges the user's real-time emotion, and the specific steps are as follows:

[0019] Based on the analysis of the STFT spectrogram by the CNN model, the time-frequency features related to emotions are further extracted;

[0020] Through the pre-trained natural language processing model, semantic information is extracted from voice commands and converted into semantic vectors;

[0021] The time-frequency features and semantic features of speech are concatenated to form a multimodal feature vector M(t), which is used as the input of the emotion recognition model.

[0022] Based on the fact that emotional states usually change dynamically over time, the second-order derivative of the multimodal feature vector is introduced To capture the acceleration changes of emotional signals;

[0023] Based on the multimodal feature vector, calculate the relative strength of each modal feature

[0024] The relative strength of each modal feature is weighted, and the fluctuation of the feature is smoothed by an exponential decay function, expressed as:

[0025]

[0026] Among them, α is the parameter that controls the decay speed, K is the number of multimodal features, and k is the index variable of the modal feature;

[0027] Based on the multimodal eigenvector M(t), a combination of second-order derivatives, normalization, and time integration is used Construct the emotional feature extraction model, the expression is:

[0028]

[0029] Where E(t) represents the emotional feature value at time t, T represents the upper limit of the time integral, ||Sk(T)|| represents the strength of the kth modal feature, and ||M(t)|| is the modulus of the overall multimodal feature vector;

[0030] The extracted emotional feature value E(t) is subjected to Z-score standardization and PCA dimensionality reduction processing;

[0031] Pair the processed emotional features with the corresponding emotional labels to form training sets, validation sets, and test sets;

[0032] Use the training set to train the classifier, and use the validation set to evaluate the performance of the model on different emotion categories and adjust the hyperparameters of the emotion feature extraction model;

[0033] After model training and validation are complete, use the test set to perform a final evaluation of the model;

[0034] According to the output of the classifier, the user's emotional state is mapped to a specific emotion category and the user's real-time emotion is output.

[0035] As a preferred solution of the soothing light intelligent control method based on voice interaction of the present invention, wherein: the corresponding light control parameters are generated, specifically, the steps are:

[0036] Through the built-in sensor interface, activate the environmental sensor to collect the external conditions of the surrounding environment;

[0037] The sensor data is converted into digital signals through A / D converters, and then denoised and formatted;

[0038] Define a mapping table between emotional states and lighting parameters, which, when mapped to high-brightness, warm-toned lighting, indicates a happy mood;

[0039] When mapped to cool-toned lighting, it indicates anxiety;

[0040] When mapped to a soft neutral light, it indicates a sad mood;

[0041] When mapped to low brightness and cool lighting, it indicates anger;

[0042] The mapped emotional state is combined with environmental data to generate control parameters for the lighting.

[0043] As a preferred solution of the method for intelligently controlling a soothing light based on voice interaction described in the present invention, the specific steps of adjusting the brightness, color temperature and color setting of the soothing light in real time are as follows:

[0044] Encapsulate the generated lighting control parameters into standard data packets and transmit them to the lighting equipment through the communication protocol;

[0045] After receiving the data packet, the lighting device returns an ACK signal to confirm successful reception. If no ACK is received, the data packet will be automatically resent until it succeeds.

[0046] Obtain the current brightness, color temperature and color value through the feedback interface of the lighting equipment;

[0047] Using the ambient light sensor and color temperature sensor, the actual current light intensity and color temperature are collected and an exponential decay function is used for smooth transition;

[0048] The actual light intensity and color temperature after smooth transition are gradually adjusted with small steps to adjust the lighting parameters.

[0049] As a preferred solution of the method for intelligently controlling a soothing lamp based on voice interaction of the present invention, the specific steps of optimizing the lighting setting in combination with environmental data are as follows:

[0050] Collect the user's environment data in real time through environmental sensors i (t);

[0051] Environmental data i (t) Perform normalization processing to generate normalized environmental data

[0052] The influence of adjusting the emotional characteristics according to the current ambient light intensity is defined as an exponential decay function γ(t) to adjust the emotional characteristic value, and the expression is:

[0053]

[0054] Among them, Llight (t) represents the light intensity value of the current environment, ||L(t)|| represents the modulus of the environmental data, and λ is the parameter that controls the attenuation speed;

[0055] Based on the emotional features E(t) and environmental data L i (t), and introduce the exponential decay function γ(t) to optimize the lighting settings, the expression is:

[0056]

[0057] Where P(t) represents the lighting control parameter at time t, i represents the index variable of the sensor data value, N represents the total number of environmental data, and β is the scaling factor;

[0058] The generated lighting control parameters P(t) are transmitted to the intelligent lighting device in real time;

[0059] After receiving the new control parameters, the lighting equipment will optimize the brightness, color temperature and color value of the light according to the parameters.

[0060] As a preferred solution of the method for intelligently controlling a soothing light based on voice interaction described in the present invention, the specific steps of continuously optimizing the light settings are as follows:

[0061] Each time the lighting control parameters are optimized, the current emotional characteristics E(t) and the current environmental data L are recorded. i (t), the current lighting setting P(t), and the user's manually adjusted lighting parameter data P user (t);

[0062] The recorded data will be stored in the database to form the state D(t) at the current time point t;

[0063] When the user manually adjusts the light, the current light parameter P(t) and the light parameter P after the user adjusts the light are calculated based on the state D(t) at time point t. user (t) the difference between;

[0064] According to the difference between the current light setting and the user's expectation, the scaling factor β and the exponential decay function γ(t) in the light parameter P(t) expression are continuously adjusted.

[0065] In a second aspect, the present invention provides a soothing light intelligent control system based on voice interaction, including a voice interaction module, a voice emotion recognition module, an emotion perception module, a light control module, a light adjustment module and a preference learning module;

[0066] The voice interaction module is used to receive the voice command input by the user through the voice interaction device, wake up the control system and start the voice emotion recognition unit;

[0067] The speech emotion recognition module is used for the speech emotion recognition unit to perform emotion analysis on the speech command, extract emotion features and judge the real-time emotion of the user;

[0068] The emotion perception module is used to enable the environment perception function based on the determined emotional state, detect the external conditions of the surrounding environment in real time, and generate corresponding lighting control parameters;

[0069] The lighting control module is used to transmit the generated lighting control parameters to the lighting control device to adjust the brightness, color temperature and color settings of the soothing light in real time;

[0070] The lighting adjustment module is used to periodically collect voice signals based on the user's continuous voice input, dynamically update the emotional state, and optimize the lighting settings in combination with environmental data;

[0071] The preference learning module is used to learn the user's emotional patterns and lighting preferences by recording the user's historical voice interaction data and environmental data, and continuously optimize the lighting settings.

[0072] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the intelligent control method of a soothing light based on voice interaction as described in the first aspect of the present invention is implemented.

[0073] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the intelligent control method for a soothing light based on voice interaction as described in the first aspect of the present invention is implemented.

[0074] The beneficial effects of the present invention are as follows: by introducing voice emotion recognition technology and combining it with the environmental perception function, the present invention can dynamically adjust the brightness, color temperature and color settings of the light according to the user's real-time emotions and surrounding environmental conditions, thereby significantly improving the user's personalized experience; by acquiring the user's emotional state through emotion recognition, the lighting effect is automatically optimized according to the user's psychological needs, and it has stronger adaptability and humanity; in addition, by continuously collecting the user's voice and environmental data, the present invention establishes a relationship between the user's emotional pattern and the lighting preference, and continuously optimizes the lighting settings through a self-learning mechanism, ensuring that the system can be adaptively adjusted as the user's needs change, making the lighting control more intelligent and personalized, and effectively improving the user's comfort and satisfaction, especially in improving emotions and relieving stress. It has obvious advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0076] Figure 1 This is a flow chart of the intelligent control method of the soothing lamp based on voice interaction in Example 1.

[0077] Figure 2 This is a flowchart of waking up the control system and starting the voice emotion recognition unit in Example 1. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0079] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0080] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0081] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a method for intelligently controlling a soothing lamp based on voice interaction, comprising the following steps:

[0082] S1. Receive the voice command input by the user through the voice interaction device, wake up the control system and start the voice emotion recognition unit.

[0083] Furthermore, the microphone array is used to beamform the voice signal, enhance the voice signal from the user's position and suppress background noise, obtain a voice signal with a high signal-to-noise ratio, and ensure high quality of basic data for subsequent processing;

[0084] Use Wiener filter combined with deep learning model to perform adaptive noise reduction on speech signals to further improve clarity;

[0085] Among them, noise is effectively suppressed and the quality of voice signals is enhanced, providing cleaner input for wake-up word detection.

[0086] Perform short-time Fourier transform (STFT) on the preprocessed speech signal to generate a speech spectrogram for feature extraction;

[0087] Use the CNN model to analyze the spectrogram, and detect the wake-up word based on the spectrogram to detect whether the wake-up word is included;

[0088] Use the CNN model to perform multi-layer convolution operations on the spectrogram to extract features related to the wake-up word;

[0089] The convolution layer extracts local features, and the pooling layer reduces the feature map. During the analysis process, an attention mechanism should be introduced to focus on the key parts related to the wake-up word.

[0090] Finally, the probability of the wake-up word is output through the fully connected layer and Softmax classifier;

[0091] If the probability of a certain wake-up word output exceeds the set threshold (such as 0.8), the system considers that the wake-up word is detected, the system switches to the wake-up state, starts the emotion recognition unit, and prepares to receive the user's voice data for emotion analysis.

[0092] S2. The speech emotion recognition unit performs emotion analysis on the speech command, extracts emotion features and determines the user's real-time emotion.

[0093] Furthermore, based on the analysis of the STFT spectrogram by the CNN model, the time-frequency features related to emotions are further extracted;

[0094] Extract semantic information from voice commands through pre-trained natural language processing models (such as BERT or GPT) and convert them into semantic vectors;

[0095] Preferably, semantic features help to identify emotional tendencies in speech, for example, by determining whether the user is expressing anger, joy, or other emotions through semantic understanding.

[0096] It should be noted that pre-trained natural language processing models (such as BERT, GPT, etc.) are existing, widely trained models that have been deeply applied to various natural language processing tasks. They are pre-trained on multiple large-scale public corpora and can effectively understand and generate natural language;

[0097] Furthermore, these models have been trained on large-scale datasets and have mastered rich language patterns and semantic information. Therefore, they can be used directly in tasks such as sentiment analysis or further optimized through fine-tuning, saving a lot of development time and resources.

[0098] The time-frequency features and semantic features of speech are concatenated to form a multimodal feature vector M(t) as the input of the emotion recognition model; through multimodal fusion, the changes in emotion at the acoustic and semantic levels can be captured at the same time.

[0099] The expression of multimodal feature vector M(t) is:

[0100] M(t)=[Stf(t),Ssem(t)];

[0101] Among them, Stf(t) is the vector representation of time-frequency features, and SSem(α) is the semantic feature vector extracted from the semantic model;

[0102] It should also be noted that the use of multimodal fusion methods combines the time-frequency features (acoustic features) and semantic features (language features) of speech signals, providing all-round information support for sentiment analysis and significantly improving the accuracy of sentiment recognition.

[0103] Based on the fact that emotional states usually change dynamically over time, the second-order derivative of the multimodal feature vector is introduced To capture the acceleration changes of emotional signals;

[0104] Compared with the first-order derivative (which reflects the instantaneous rate of change of the signal), the second-order derivative can describe the dramatic fluctuation of the emotional signal. For example:

[0105] First-order derivative: represents the rate of change of the emotion signal over time (such as an increase or decrease in emotion intensity).

[0106] Second-order derivative: represents the acceleration of the change of the emotional signal (i.e., the intensity of the change in emotional intensity).

[0107] Therefore, the second-order derivative It is an important means to capture emotional fluctuations, and this derivative is used to measure the dynamic intensity of emotional signals.

[0108] Based on the multimodal feature vector, calculate the relative strength of each modal feature Measure the relative contribution of each modality feature to the sentiment feature;

[0109] The relative strength of each modal feature is weighted, and the fluctuation of the feature is smoothed by an exponential decay function, expressed as:

[0110]

[0111] Among them, α is a parameter that controls the attenuation speed and is used to adjust the smoothness of emotional fluctuations. A larger α will make the system more sensitive to emotional fluctuations, while a smaller α will smooth fluctuations and make the system react more slowly to emotional changes. K is the number of multimodal features, and k is the index variable of the modal feature;

[0112] Based on the multimodal eigenvector M(t), a combination of second-order derivatives, normalization, and time integration is used Construct the emotional feature extraction model, the expression is:

[0113]

[0114] Among them, E(t) represents the emotional feature value at time t, reflecting the emotional intensity, T represents the upper limit of the time integral, ||Sk(t)|| represents the intensity of the kth modal feature (vector modulus), ||M(t)|| is the modulus of the overall multimodal feature vector, representing the overall intensity of the multimodal feature;

[0115] The extracted emotional feature value E(t) is subjected to Z-score standardization and PCA (principal component analysis) dimensionality reduction processing to reduce the dimension and improve performance and computing efficiency;

[0116] Pair the processed emotional features with the corresponding emotional labels (such as anger, joy, and sadness) to form training sets, validation sets, and test sets;

[0117] Use the training set to train the classifier so that it can predict the correct emotion category based on the input features;

[0118] It should be noted that commonly used classifiers include:

[0119] Traditional machine learning models: such as support vector machine (SVM), random forest, K-nearest neighbor (KNN), etc.

[0120] Deep learning models: such as multi-layer perceptron (MLP), convolutional neural network (CNN), long short-term memory network (LSTM), etc.

[0121] During the training process, the classifier continuously adjusts the weights and biases of the model to minimize the error between the output emotion category prediction and the true label.

[0122] Evaluate the performance of the model on different emotion categories through the validation set, adjust the hyperparameters of the emotion feature extraction model, and optimize the classification effect, for example:

[0123] Classifier hyperparameters: such as the kernel function type in SVM, the learning rate in deep learning models, the number of hidden layer units, etc.

[0124] Emotional feature extraction hyperparameters: such as the number of principal components in PCA, the window size in Z-score standardization, etc.

[0125] Through cross-validation (k-fold cross-validation), the best hyperparameter combination is found to optimize the classification effect of the model.

[0126] After model training and validation are complete, the model is finally evaluated using the test set to measure the model's performance on completely unseen data;

[0127] According to the output of the classifier, the user's emotional state is mapped to a specific emotion category, the user's real-time emotion is output, and corresponding feedback is provided according to the application scenario.

[0128] S3. Based on the determined emotional state, the environment perception function is enabled to detect the external conditions of the surrounding environment in real time and generate corresponding lighting control parameters.

[0129] Furthermore, environmental sensors are activated through the built-in sensor interface, including but not limited to: light sensor: detects the natural light intensity of the current environment; color temperature sensor: detects the color temperature of the existing light source; temperature and humidity sensor: detects the temperature and humidity of the environment to assist in lighting adjustment; time sensor: obtains the current time to determine whether it is day or night; location sensor (such as GPS or indoor positioning) is used to determine the user's physical environment, such as whether at home, in the office or other places.

[0130] The sensor data is converted into digital signals through A / D converters, and then denoised and formatted;

[0131] De-noising specifically refers to using a median filter or a Kalman filter to denoise the collected data.

[0132] Data formatting specifically refers to standardizing the data and aligning the timestamps to ensure that all sensor data are synchronized.

[0133] A mapping table defining emotional states and lighting parameters, which, when mapped to high brightness (ranging from 70% to 100%) and warm tones (e.g., 2700K to 3000K) lights, indicates a happy mood;

[0134] When mapped to cool-toned lights (e.g., 5000K to 6500K), it helps users relax and indicates anxiety;

[0135] When mapped to soft neutral light (such as 4000K), it avoids excessive stimulation and indicates sad emotions;

[0136] When mapped to low brightness and cool-toned lighting, it helps users calm down and express anger;

[0137] Among them, the more specific standard for low brightness is 10%-30%, that is, the brightness of the light is adjusted between 10% and 30% of the total brightness, which is used to ensure that the light is not too bright and can achieve the effect of soothing or easing emotions.

[0138] The standard for cool colors is 5500K to 6500K. This range is usually defined as cool colors. It can give people a cool and calm feeling and help calm emotions.

[0139] It should be noted that the specific steps of defining the mapping table between emotional states and lighting parameters are as follows:

[0140] First, we need to clarify the classification of emotions and divide them into several common emotional states. Common emotional classifications include: happiness, anxiety, sadness, anger, calmness, and fatigue;

[0141] Secondly, determine that the lighting parameters can be adjusted through the following dimensions to suit different emotional states: Brightness: Indicates the intensity of the light, usually expressed as a percentage or lumen value (0-100%); Color temperature (CCT, Correlated Color Temperature): Indicates the hue of the light, in Kelvin (K) (the common range is 2700K to 6500K); Color (RGB value): If the lamp supports RGB mode, the color of the light can be adjusted through the RGB value;

[0142] Combining human physiological and psychological responses to light, we can establish a preliminary relationship between emotions and lighting parameters. For example:

[0143] Pleasant mood: prefers high brightness and warm tones to create a comfortable and positive atmosphere.

[0144] Anxiety: Cool colors and medium brightness can help users relax and reduce anxiety.

[0145] Sad mood: Soft neutral light (medium brightness, moderate color temperature) can avoid excessive stimulation and bring a quiet and peaceful feeling.

[0146] Angry emotions: Low-brightness, cool-toned lighting can help calm emotions and avoid further stimulation.

[0147] According to the characteristics of each emotion, determine the specific range of lighting parameters. The specific operations are as follows:

[0148] Brightness range: Set the brightness percentage according to different emotions (such as 0-100%).

[0149] For example, when the emotional state is "happy", the initially generated brightness is 80%, and when the ambient light is strong, the brightness is corrected to -20%, and the final brightness is 60%.

[0150] Color temperature range: Set the color temperature range according to different emotions (such as 2700K-6500K).

[0151] For example, if the emotional state is "anxiety", the initially generated color temperature is 6000K, the current time is night, the color temperature is corrected to -500K, and the final color temperature is 5500K.

[0152] Color (RGB value): If the lamp supports RGB mode, you can define different RGB values ​​according to your mood.

[0153] For example, when the emotional state is "angry", the initially generated color is cold blue (RGB: 135, 206, 250), and when the ambient temperature is high, the final color is adjusted to a slightly cooler blue.

[0154] Establish a mapping relationship between the emotional state and the specific values ​​of the lighting parameters to form a table, as shown below:

[0155] Emotional state Brightness range (%) Color temperature range(K) RGB color value Pleasure 70％—100％ 2700K—3000K (255,223,186) (warm orange) anxiety 40％—70％ 5000K—6500K (173,216,230)(light blue) sad 30％—50％ 4000K (144,164,174)(gray-blue) anger 10％—30％ 5500K—6500K (135,206,250)(cold blue)

[0156] Note: RGB color values ​​can be adjusted according to the color range supported by the actual lamp.

[0157] After the mapping table is established, the system will dynamically adjust the lighting parameters according to real-time environmental data (such as the current natural light intensity, time, temperature and humidity, etc.). The specific adjustment methods are as follows:

[0158] Natural light intensity: If the ambient light is bright, the system can automatically reduce the brightness of the lights.

[0159] Time of day factor: If it is night time, the color temperature will shift towards warmer tones.

[0160] Temperature and humidity influence: In high temperature or high humidity environment, the system will give priority to cool colors.

[0161] As the interaction between users and the system increases, the system can gradually learn the user's personalized needs by recording the user's lighting preferences and mood changes, and optimize the mapping table between emotions and lighting parameters to make it more in line with personal habits.

[0162] Combine the mapped emotional state with environmental data to generate control parameters for lighting;

[0163] It should be noted that the emotional state and environmental data are combined to generate the lighting control parameters. Here, "combination" means that the system corrects and optimizes the preliminary mapping value of the emotional state in combination with the environmental data. The specific steps are as follows:

[0164] Weight assignment: Assign different weights to emotional state and environmental data. For example, in some cases, emotional state may be more important than environmental data, and vice versa. The weights can be adjusted according to the actual usage scenario (such as 70% for emotion and 30% for environmental data). This weight can be dynamically adjusted through user feedback or machine learning.

[0165] Formula combination: weighted calculation of the preliminary lighting parameters (brightness, color temperature, color) of the emotional state and the environmental data. For example, a simple combination formula is as follows:

[0166] Final brightness = w 情绪 ×Emotion brightness value + w 环境 × Ambient brightness correction value;

[0167] Among them, w 情绪 and w 环境 is the weight of the emotional state and the environmental data (such as 0.7 and 0.3), the emotional brightness value represents the initial brightness value obtained from the emotional state mapping table, and the environmental brightness correction value represents the brightness correction value based on the ambient light data. For example, if the current ambient light is very strong, the system may reduce the light brightness; similarly, the generation of color temperature and color value will also be weighted synthesis based on the initial value of the emotional state and the environmental data;

[0168] S4. The generated lighting control parameters are transmitted to the lighting control device to adjust the brightness, color temperature and color settings of the soothing light in real time.

[0169] Furthermore, the generated lighting control parameters (brightness, color temperature, color value) are encapsulated into a standard data packet JSON (JavaScript Object Notation). The specific data format is as follows:

[0170] Brightness (L): 0-100 percentage; Color temperature (CCT): in Kelvin, ranging from 2700K≤CCT≤6500K; Color (RGB): if color adjustment is supported, it is expressed using RGB values, ranging from 0≤R / G / B≤255.

[0171] Data packets are transmitted to lighting equipment via communication protocols (such as Wi-Fi, Zigbee, Bluetooth) to ensure device compatibility and transmission reliability;

[0172] After receiving the data packet, the lighting device returns an ACK signal to confirm successful reception. If no ACK is received, the data packet will be automatically resent until it succeeds.

[0173] Obtain the current brightness, color temperature and color value (such as RGB value) through the feedback interface of the lighting equipment;

[0174] Using the ambient light sensor and color temperature sensor, the actual current light intensity and color temperature are collected and an exponential decay function is used for smooth transition;

[0175] The actual light intensity and color temperature after smooth transition are gradually adjusted with small steps to ensure smooth and natural changes. The specific adjustments are as follows:

[0176] Brightness adjustment: gradually increase or decrease the brightness to make it gradually approach the target brightness.

[0177] Color temperature adjustment: gradually adjust the color temperature to simulate the changing rhythm of natural light.

[0178] Color adjustment: By adjusting the proportion of RGB channels, the color gradually changes to the target value.

[0179] It should be noted that after each adjustment, it is necessary to wait for a predefined time interval (e.g., 100 milliseconds) before making the next adjustment;

[0180] During the adjustment process, sensors are used for real-time monitoring to ensure that lighting changes are in line with expectations, such as:

[0181] Light sensor feedback: The system detects changes in the current ambient light through the light sensor to confirm whether the brightness gradually reaches the target value.

[0182] Color temperature sensor feedback: The color temperature sensor monitors color temperature changes in real time to ensure that the color temperature adjustment process meets expectations.

[0183] Color sensor or device feedback: If the fixture supports color feedback, the system will confirm whether the RGB color adjustment is accurate through the device self-test function or color sensor.

[0184] Furthermore, to ensure that the lighting adjustment has been successful, the system will re-acquire the sensor data after each adjustment cycle and compare it with the target value:

[0185] Brightness verification: Compare the current brightness with the target brightness. If the deviation between the two is within an acceptable range (such as ±5%), the adjustment is considered complete; otherwise, continue to adjust.

[0186] Color temperature verification: Detect the difference between the current color temperature and the target color temperature. If the deviation is within the specified range (such as ±200K), the adjustment is successful; otherwise, continue to adjust.

[0187] Color verification: compare the current RGB value with the target RGB value. If the difference does not exceed the threshold (such as ±10 units), the color adjustment is completed; otherwise, continue to adjust.

[0188] If the lighting parameters do not reach the target value within the specified adjustment time (e.g. 5 seconds), the system will automatically resend the control command and the adjustment process will be repeated until the lighting parameters are consistent with the target value.

[0189] Resend mechanism: The system will determine whether to resend the lighting control command based on the feedback data. If the deviation between the brightness, color temperature or color and the target value exceeds the acceptable range, the system will automatically recalculate the difference and send a new control command.

[0190] Dynamic Adjustment: When the mood or environment changes, the system will automatically regenerate new lighting parameters and make new adjustments according to the above process.

[0191] When all lighting parameters are adjusted to the user's preset target values ​​and the feedback is verified, the system will end the adjustment process and enter standby mode to prepare for the next adjustment. During this period, the system will regularly check the emotional state and environmental data to ensure that the lighting settings always match the user's needs.

[0192] S5. Based on the user's continuous voice input, voice signals are collected regularly, emotional status is dynamically updated, and lighting settings are optimized in combination with environmental data.

[0193] Furthermore, the environmental sensor collects the user's environmental data in real time. i (t);

[0194] Environmental data i (t) Perform normalization processing to generate normalized environmental data The expression is:

[0195]

[0196] in, Represents the raw data L of the i-th sensor i (t) is the normalized value, i represents the index variable of the sensor data value, N represents the total number of environmental data, and ||L(t)|| represents the modulus of the environmental data;

[0197] The modulus length of environmental data ||L(t)|| is expressed as:

[0198]

[0199] It should be noted that this normalization process ensures that the relative weights of different environmental data can be reasonably expressed and prevents the absolute value of a certain environmental data from being too large, resulting in excessive control over the lighting settings.

[0200] The influence of emotional characteristics is adjusted according to the current ambient light intensity. The exponential decay function γ(t) is defined to adjust the emotional characteristic value, so that the influence of emotional characteristics on lighting settings is smoother. In particular, when the ambient light intensity is large, the role of emotional characteristics in lighting control is weakened. The expression is:

[0201]

[0202] Among them, L light (t) represents the light intensity value of the current environment, ||L(t)|| represents the modulus of the environmental data, and λ is a parameter that controls the attenuation speed. A larger λ indicates a stronger attenuation, that is, the influence of emotional characteristics on light brightness will be weakened in a strong light environment;

[0203] Based on the emotional features E(t) and environmental data L i (t), and introduce the exponential decay function γ(t) to optimize the lighting settings, the expression is:

[0204]

[0205] Where P(t) represents the lighting control parameters at time t, including the brightness, color temperature and color value of the light, i represents the index variable of the sensor data value, N represents the total number of environmental data, and β is the scaling factor, which is used to adjust the brightness and color intensity of the light as a whole;

[0206] The generated lighting control parameters P(t) are transmitted to the intelligent lighting device in real time;

[0207] After receiving the new control parameters, the lighting equipment will optimize the brightness, color temperature and color value of the light according to the parameters to prevent drastic changes in the light from having a negative impact on the user experience;

[0208] S6. By recording the user's historical voice interaction data and environmental data, learn the user's emotional patterns and lighting preferences, and continuously optimize the lighting settings.

[0209] Furthermore, each time the lighting control parameters are optimized, the current emotional characteristics E(t) and the current environmental data L are recorded. i (t), the current lighting setting P(t), and the user's manually adjusted lighting parameter data P user (t);

[0210] The recorded data will be stored in the database to form the state D(t) at the current time point t;

[0211] The record items for each interaction are:

[0212] D(t)={E(t),L i (t),P(t),P user(t)};

[0213] Among them, P user (t) represents the lighting control parameters after manual adjustment by the user;

[0214] When the user manually adjusts the light, the current light parameter P(t) and the light parameter P after the user adjusts the light are calculated based on the state D(t) at time point t. user The difference between (t) is expressed as:

[0215] ΔP(t)=P user (t)-P(t);

[0216] Among them, ΔP(t) represents the difference between the current lighting setting and the user's expectation. The system needs to update the lighting control strategy based on this difference.

[0217] According to the difference between the current lighting setting and the user's expectation, the scaling factor β and the exponential decay function γ(t) in the lighting parameter P(t) expression are continuously adjusted;

[0218] It should be noted that if the user prefers to increase or decrease the overall light brightness or color temperature, the scaling factor β in the light parameter P(t) expression needs to be adjusted, as shown in the following expression:

[0219] β t+1 =β t +η β ΔP(t)

[0220] Among them, β t+1 represents the scaling factor at the next time step t+1, β t represents the scaling factor at time step t, η β represents the learning rate used to update the scaling factor β;

[0221] If the user's lighting setting preferences are consistent or different under different ambient light intensities, the exponential decay function γ(t) in the expression of the lighting parameter P(t) needs to be adjusted, and the expression is:

[0222] γ t+1 (t) = γ t (t)+η γ ΔP(t);

[0223] Among them, γ t+1 (t) represents the exponential decay function after the next time step t+1, γ t (t) exponential decay function at time step t;

[0224] By adjusting β and γ(t), the user’s lighting preferences under different emotional and environmental conditions can be gradually learned, making the lighting settings more in line with user expectations.

[0225] The present embodiment also provides a soothing light intelligent control system based on voice interaction, including: a voice interaction module, a voice emotion recognition module, an emotion perception module, a light control module, a light adjustment module and a preference learning module; the voice interaction module is used to receive voice commands input by the user through a voice interaction device, wake up the control system and start the voice emotion recognition unit; the voice emotion recognition module is used to perform emotion analysis on the voice command by the voice emotion recognition unit, extract emotion features and judge the real-time emotion of the user; the emotion perception module is used to enable the environmental perception function based on the judged emotional state, detect the external conditions of the surrounding environment in real time, and generate corresponding light control parameters; the light control module is used to transmit the generated light control parameters to the light control device, and adjust the brightness, color temperature and color settings of the soothing light in real time; the light adjustment module is used to regularly collect voice signals based on the user's continuous voice input, dynamically update the emotional state, and optimize the light settings in combination with the environmental data; the preference learning module is used to learn the user's emotional pattern and light preference by recording the user's historical voice interaction data and environmental data, and continuously optimize the light settings.

[0226] This embodiment also provides a computer device, which is suitable for the case of a soothing light intelligent control method based on voice interaction, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the soothing light intelligent control method based on voice interaction as proposed in the above embodiment.

[0227] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0228] The present embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent control method for a soothing lamp based on voice interaction as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage device, a flash memory, a disk or an optical disk.

[0229] In summary, the present invention introduces voice emotion recognition technology and combines it with the environmental perception function to dynamically adjust the brightness, color temperature and color settings of the lights according to the user's real-time emotions and surrounding environmental conditions, thereby significantly improving the user's personalized experience; the user's emotional state is acquired through emotion recognition, thereby automatically optimizing the lighting effect according to the user's psychological needs, which has stronger adaptability and humanity; in addition, the present invention continuously collects the user's voice and environmental data, establishes the association between the user's emotional pattern and the lighting preference, and continuously optimizes the lighting settings through a self-learning mechanism, ensuring that the system can be adaptively adjusted as user needs change, making lighting control more intelligent and personalized, effectively improving the user's comfort and satisfaction, especially in improving emotions and relieving stress.

[0230] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a soothing lamp intelligent control method based on voice interaction is provided.

[0231] The experiment was conducted in a typical smart home environment. The experimental scenario included users interacting with the soothing light through voice interaction devices. The devices used in the experiment included:

[0232] Microphone array: used to capture the user's voice input and enhance the signal through beamforming technology.

[0233] Environmental sensor: used to collect ambient conditions, including light intensity, temperature, humidity, etc.

[0234] Smart lighting equipment: It has the function of adjusting brightness, color temperature and color, and adjusts lighting settings by receiving control parameters through the protocol.

[0235] Speech processing and emotion recognition system: Perform sentiment analysis on voice commands based on CNN and NLP models to extract the user's emotional characteristics.

[0236] First, the user's voice signal is captured through the microphone array, and the signal is processed for noise reduction through the deep learning model. Then, the CNN model is used to analyze the voice signal and detect the wake-up word. After the system is successfully awakened, it enters the emotion recognition mode.

[0237] At this time, the speech signal is transformed into a spectrum through short-time Fourier transform, and combined with natural language processing technology to extract semantic information and generate a multimodal feature vector. By weighting the multimodal features and calculating the second-order derivative, the emotion recognition module can capture the user's emotional changes.

[0238] Furthermore, after emotion recognition is completed, the system generates corresponding lighting control parameters based on the user's emotional state. In the experiment, a variety of mappings between emotions and lighting parameters were defined, such as high brightness and warm light for joyful emotions, cool light for anxious emotions, soft neutral light for sad emotions, and low brightness and cool light for angry emotions. The system transmits these parameters to the smart lighting device through the communication protocol, and the device adjusts the lighting settings according to the instructions.

[0239] In order to ensure the comprehensiveness of the experiment, multiple control groups were set up to use the traditional manual control-based lighting adjustment scheme as a comparison. Each group of experiments conducted multiple emotional state tests and lighting adjustments, and continuously recorded the user's feedback, emotional changes, and lighting settings.

[0240] The details are shown in Table 1 below:

[0241] Table 1 Experimental data comparison table

[0242]

[0243] It can be seen from the above data that under different emotional states, there are certain differences between the lighting control parameters automatically generated by the system and the parameters manually adjusted by the user.

[0244] Under pleasant emotions, the system generates a light brightness of 800 lumens and a color temperature of 3000K. After manual adjustment by the user, the brightness is slightly reduced to 750 lumens. This shows that the brightness parameters generated by the system are basically consistent with the user's expectations, and warm light helps maintain the comfort of pleasant emotions. This result proves that lighting control based on emotion recognition can effectively capture the user's emotional needs and generate reasonable lighting parameters.

[0245] Under anxiety, the system generates a light brightness of 500 lumens and a color temperature of 6000K. After manual adjustment by the user, the brightness is slightly reduced to 480 lumens. Cool-toned lights have a certain effect in relieving anxiety, which matches the system's automatically generated parameters. Compared with traditional manual control methods, intelligent lighting control based on emotion recognition can respond to user emotional changes more quickly and accurately, reducing the number of manual adjustments by users.

[0246] For the sad emotion, the system generated a brightness of 400 lumens and a color temperature of 4500K. After manual adjustment by the user, the brightness was slightly increased to 410 lumens. The soft neutral light performed well in relieving sad emotions, and the manual adjustment range of the user was small, indicating that the lighting parameters generated by the system were consistent with the user's expectations.

[0247] Under the angry emotion, the system generates a brightness of 300 lumens and a color temperature of 5000K. The user manually lowers the brightness slightly to 290 lumens. Low-brightness, cool-colored lights help alleviate anger, and the lighting settings executed by the system are basically consistent with the manual adjustment results of the user.

[0248] From the overall data analysis, the intelligent control method of soothing lights based on voice interaction and emotion recognition has shown significant advantages in capturing user emotions and generating corresponding lighting control parameters. Compared with the traditional manual adjustment method, the system can adjust the lighting settings in real time and dynamically, reducing the user's operating burden and showing higher accuracy in emotional matching.

[0249] In addition, the experimental results show that the system has good adaptability in the linkage between emotion recognition and lighting control, which significantly improves the user experience. By combining environmental data with the user's interaction history, the system can continuously optimize itself and gradually learn the user's lighting preferences, making the lighting settings more personalized and intelligent.

[0250] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A soothing light intelligent control method based on voice interaction, characterized in that: include, Receive the voice command input by the user through the voice interaction device, wake up the control system and start the voice emotion recognition unit; The speech emotion recognition unit performs emotion analysis on the speech commands, extracts emotion features and determines the user's real-time emotions; Based on the determined emotional state, the environmental perception function is enabled to detect the external conditions of the surrounding environment in real time and generate corresponding lighting control parameters; The generated lighting control parameters are transmitted to the lighting control device to adjust the brightness, color temperature and color settings of the soothing light in real time; Based on the user's continuous voice input, voice signals are collected regularly, emotional status is dynamically updated, and lighting settings are optimized in combination with environmental data; By recording the user's historical voice interaction data and environmental data, we learn the user's emotional patterns and lighting preferences and continuously optimize the lighting settings.

2. The intelligent control method of the soothing light based on voice interaction according to claim 1, characterized in that: The specific steps of receiving the voice command input by the user through the voice interaction device, waking up the control system and starting the voice emotion recognition unit are as follows: The microphone array is used to beamform the voice signal, enhance the direction of the user's voice, and obtain a voice signal with a high signal-to-noise ratio. Use Wiener filter combined with deep learning model to perform adaptive noise reduction on speech signals; Perform short-time Fourier transform on the preprocessed speech signal to generate an STFT spectrum of the speech; The CNN model is used to analyze the STFT spectrum graph to detect whether it contains the wake-up word. When the wake-up word is detected successfully, the system switches to the wake-up state and starts the emotion recognition unit.

3. The intelligent control method of the soothing lamp based on voice interaction according to claim 2, characterized in that: The speech emotion recognition unit performs emotion analysis on the speech command, extracts emotion features and determines the user's real-time emotion. The specific steps are: Based on the analysis of the STFT spectrogram by the CNN model, the time-frequency features related to emotions are further extracted; Through the pre-trained natural language processing model, semantic information is extracted from voice commands and converted into semantic vectors; The time-frequency features and semantic features of speech are concatenated to form a multimodal feature vector M(t), which is used as the input of the emotion recognition model. Based on the fact that emotional states usually change dynamically over time, the second-order derivative of the multimodal feature vector is introduced To capture the acceleration changes of emotional signals; Based on the multimodal feature vector, calculate the relative strength of each modal feature The relative strength of each modal feature is weighted, and the fluctuation of the feature is smoothed by an exponential decay function, expressed as: Among them, α is the parameter that controls the decay speed, K is the number of multimodal features, and k is the index variable of the modal feature; Based on the multimodal eigenvector M(t), a combination of second-order derivatives, normalization, and time integration is used Construct a sentiment feature extraction model, the expression is: Among them, E(t) represents the emotional feature value at time t, T represents the upper limit of the time integral, and ||S k (t)|| represents the strength of the kth modal feature, and ||M(t)|| is the modulus of the overall multimodal feature vector; The extracted emotional feature value E(t) is subjected to Z-score standardization and PCA dimensionality reduction processing; Pair the processed emotional features with the corresponding emotional labels to form training sets, validation sets, and test sets; Use the training set to train the classifier, and use the validation set to evaluate the performance of the model on different emotion categories and adjust the hyperparameters of the emotion feature extraction model; After model training and validation are complete, use the test set to perform a final evaluation of the model; According to the output of the classifier, the user's emotional state is mapped to a specific emotion category and the user's real-time emotion is output.

4. The intelligent control method of the soothing lamp based on voice interaction according to claim 3, characterized in that: The specific steps of generating corresponding lighting control parameters are as follows: Through the built-in sensor interface, activate the environmental sensor to collect the external conditions of the surrounding environment; The sensor data is converted into digital signals through A / D converters, and then denoised and formatted; Define a mapping table between emotional states and lighting parameters, which, when mapped to high-brightness, warm-toned lighting, indicates a happy mood; When mapped to cool-toned lighting, it indicates anxiety; When mapped to a soft neutral light, it indicates a sad mood; When mapped to low brightness and cool lighting, it indicates anger; The mapped emotional state is combined with environmental data to generate control parameters for the lighting.

5. The intelligent control method of the soothing lamp based on voice interaction according to claim 4, characterized in that: The specific steps of adjusting the brightness, color temperature and color setting of the soothing light in real time are as follows: Encapsulate the generated lighting control parameters into standard data packets and transmit them to the lighting equipment through the communication protocol; After receiving the data packet, the lighting device returns an ACK signal to confirm successful reception. If no ACK is received, the data packet will be automatically resent until it succeeds. Obtain the current brightness, color temperature and color value through the feedback interface of the lighting equipment; Using the ambient light sensor and color temperature sensor, the actual current light intensity and color temperature are collected and an exponential decay function is used for smooth transition; The actual light intensity and color temperature after smooth transition are gradually adjusted with small steps to adjust the lighting parameters.

6. The intelligent control method of a soothing lamp based on voice interaction according to claim 5, characterized in that: The specific steps of optimizing the lighting settings in combination with the environmental data are as follows: Collect the user's environment data in real time through environmental sensors i (t); Environmental data i (t) Perform normalization processing to generate normalized environmental data The influence of adjusting the emotional characteristics according to the current ambient light intensity is defined as an exponential decay function γ(t) to adjust the emotional characteristic value, and the expression is: Among them, L light (t) represents the light intensity value of the current environment, ||L(t)|| represents the modulus of the environmental data, and λ is the parameter that controls the attenuation speed; Based on the emotional features E(t) and environmental data L i (t), and introduce the exponential decay function γ(t) to optimize the lighting settings, the expression is: Where P(T) represents the lighting control parameter at time t, i represents the index variable of the sensor data value, N represents the total number of environmental data, and β is the scaling factor; The generated lighting control parameters P(t) are transmitted to the intelligent lighting device in real time; After receiving the new control parameters, the lighting equipment will optimize the brightness, color temperature and color value of the light according to the parameters.

7. The intelligent control method of a soothing lamp based on voice interaction according to claim 6, characterized in that: The specific steps of continuously optimizing the lighting settings are as follows: Each time the lighting control parameters are optimized, the current emotional characteristics E(t) and the current environmental data L are recorded. i (t), the current lighting setting P(t), and the user's manually adjusted lighting parameter data P user (t); The recorded data will be stored in the database to form the state D(t) at the current time point t; When the user manually adjusts the light, the current light parameter P(t) and the light parameter P after the user adjusts the light are calculated based on the state D(t) at time point t. user (t) the difference between; According to the difference between the current light setting and the user's expectation, the scaling factor β and the exponential decay function γ(t) in the light parameter P(t) expression are continuously adjusted.

8. A soothing light intelligent control system based on voice interaction, based on the soothing light intelligent control method based on voice interaction according to any one of claims 1 to 7, characterized in that: Including voice interaction module, voice emotion recognition module, emotion perception module, lighting control module, lighting adjustment module and preference learning module; The voice interaction module is used to receive the voice command input by the user through the voice interaction device, wake up the control system and start the voice emotion recognition unit; The speech emotion recognition module is used for the speech emotion recognition unit to perform emotion analysis on the speech command, extract emotion features and judge the real-time emotion of the user; The emotion perception module is used to enable the environment perception function based on the determined emotional state, detect the external conditions of the surrounding environment in real time, and generate corresponding lighting control parameters; The lighting control module is used to transmit the generated lighting control parameters to the lighting control device to adjust the brightness, color temperature and color settings of the soothing light in real time; The lighting adjustment module is used to periodically collect voice signals based on the user's continuous voice input, dynamically update the emotional state, and optimize the lighting settings in combination with environmental data; The preference learning module is used to learn the user's emotional patterns and lighting preferences by recording the user's historical voice interaction data and environmental data, and continuously optimize the lighting settings.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent control method of the soothing lamp based on voice interaction described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent control method of a soothing lamp based on voice interaction described in any one of claims 1 to 7 are implemented.

Images