Intelligent light language generation method and system

Abstract

The invention discloses an intelligent light language generation method and system, and the method comprises the steps: obtaining the emotion expression information of a user in advance, and generating an emotion demand tag; based on the emotion demand label, light language parameters are generated, and the light language parameters comprise the color, the brightness, the frequency and the rhythm of light; outputting a light language based on the light language parameter in combination with the hardware information and the context awareness information; and based on the feedback signal of the user on the light language, adjusting the light language parameter so as to output a new light language. The emotion demand label is generated based on the user emotion expression information, the personalized light language is output in combination with the hardware information and the context awareness information, and visual emotion transmission and a comfortable atmosphere are provided. Light language parameters are continuously optimized according to user feedback, accurate service is ensured, the self-adaptive evolution ability is achieved, user habits are learned for a long time, real-time adjustment is achieved, the limitation of a traditional method is broken through, and the man-machine interaction efficiency and quality are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical language generation, and in particular relates to an optical language intelligent generation method and system. Background Art

[0002] In modern human-computer interaction, although voice, tactile feedback and image display are the main ways of communication, these traditional means of interaction have certain limitations when facing diverse scenarios and complex needs. Especially in terms of emotional expression, voice and text are often difficult to accurately convey delicate emotions, while light signals, with their color, brightness and dynamic changes, can more intuitively convey emotions and create a comfortable or warning atmosphere. In addition, in noisy environments or when facing hearing-impaired users, traditional voice interaction may fail. At this time, light language can provide silent and clear feedback to ensure the effective communication of information.

[0003] Traditional light language generation methods mainly rely on predefined, fixed signal designs to convey information or create an atmosphere. These methods usually have the following defects:

[0004] Traditional light language generation methods generally use static light language expression, and the light signals are usually pre-set, which cannot be flexibly adjusted according to changes in user emotions and needs, and cannot meet the diverse and personalized needs of users. For example, in a specific scenario, each user's emotional fluctuations and reactions are unique, and traditional methods are difficult to provide targeted support.

[0005] In addition, traditional methods are usually developed for specific hardware platforms, and different devices (such as smart lamps, mobile phone screens, and wearable devices) have different display capabilities and performance limitations. Traditional methods find it difficult to maintain a consistent and personalized light language experience on multiple platforms. Summary of the invention

[0006] The present invention provides a method and system for intelligently generating light language, so as to solve the problem that the traditional light language generation method adopts static light language expression, resulting in the inability to flexibly adjust according to changes in user emotions and needs, the problem that the light language effect is difficult to maintain consistency on multiple platforms due to development for a specific hardware platform, and the problem that the light language effect is not optimized enough due to the inability to adjust the light according to environmental parameters.

[0007] The technical solution adopted by the present invention is:

[0008] A light language intelligent generation method, which pre-acquires user emotion expression information and generates emotion demand labels, the method also includes:

[0009] Based on the emotional demand tag, generating light language parameters, the light language parameters including light color, brightness, frequency, and rhythm;

[0010] Based on the light language parameters, combined with the hardware information and the situational awareness information, output the light language;

[0011] Based on the user's feedback signal on the light language, the light language parameters are adjusted to output a new light language.

[0012] Obtain user emotional expression information in advance and generate emotional demand labels, specifically:

[0013] The user's emotional expression information includes at least one of language, text, facial expression, and biological data;

[0014] Extracting feature parameters corresponding to the user's emotional expression information, generating a feature vector including at least one of the feature parameters, and generating an emotional demand label based on a deep learning model;

[0015] The emotional need labels include relaxation, anxiety, or excitement.

[0016] Extract the characteristic parameters corresponding to the user's emotional expression information, specifically:

[0017] The biological data includes heart rate data or respiration data, and the heart rate variability corresponding to the heart rate data and the respiration frequency corresponding to the respiration data are calculated to generate characteristic parameters corresponding to the biological data;

[0018] Extracting sound features corresponding to the language, the sound features including fundamental frequency, energy or zero-crossing rate, to generate feature parameters corresponding to the language;

[0019] Applying natural language processing technology to perform part-of-speech tagging or sentiment classification on the text to generate feature parameters corresponding to the text;

[0020] Based on a computer vision algorithm, facial features that can map emotion categories are extracted to generate feature parameters corresponding to the facial expressions.

[0021] Based on the emotional demand label, light language parameters are generated, specifically:

[0022] Establishing a mapping relationship between the emotional demand label and the light language parameter, and setting initial light language parameters for the emotional demand label one by one;

[0023] The historical data is called up to analyze the user's usage habits, and the mapping relationship is adjusted according to the user's usage habits.

[0024] Based on the light language parameters, combined with the hardware information and the situational awareness information, the light language is output, specifically:

[0025] According to a preset hardware fingerprint library, the connected hardware device is identified, the corresponding hardware information is obtained, and the optical language parameters are adjusted according to a standardized optical language parameter library covering a variety of hardware information;

[0026] By using a variety of sensors, environmental data is collected in real time, combined with the emotional demand tags, to form situational awareness information, and the light language parameters are adjusted according to the situational awareness information;

[0027] The light language parameters are transmitted to a connected hardware device so as to output the light language through the hardware device.

[0028] According to a standardized optical language parameter library covering a variety of hardware information, the optical language parameters are adjusted, specifically:

[0029] The hardware information includes color modes, light source intensity, and performance parameters supported by the hardware device;

[0030] Adjust the color parameters in the light language parameters according to the color mode supported by the hardware device;

[0031] Adjust the brightness parameter in the light language parameters according to the light intensity of the hardware device;

[0032] According to the performance parameters of the hardware equipment, adjust the frequency parameters and rhythm parameters in the light language parameters.

[0033] Through a variety of sensors, environmental data is collected in real time, combined with the emotional demand tags, to form situational awareness information, and the light language parameters are adjusted according to the situational awareness information, specifically:

[0034] The environmental data includes ambient light, temperature and humidity.

[0035] According to the environmental data, combined with the emotional demand label, situational awareness information is formed,

[0036] According to the ambient light, adjust the brightness of the light language parameter corresponding to the emotional demand label;

[0037] According to the ambient temperature and humidity, the color in the light language parameter corresponding to the emotional demand label is adjusted.

[0038] Based on the user's feedback signal on the light language, the light language parameters are adjusted to output a new light language, specifically:

[0039] The feedback signal includes user emotion expression information,

[0040] Collecting the user's emotional expression information in real time during the light language output process, and adjusting the light language parameters according to the trend change of the user's emotional expression information;

[0041] and / or, the feedback signal includes a user's satisfaction evaluation,

[0042] The light language parameters are adjusted according to the user's satisfaction evaluation.

[0043] According to the feedback signal, the mapping relationship between the emotion demand label and the light language parameter is iterated.

[0044] The present invention also provides a light language intelligent generation system, comprising:

[0045] Emotional demand generation module, used to obtain user emotional expression information and generate emotional demand labels;

[0046] A light language parameter generating module, used to generate light language parameters based on the emotional demand label, and adjust the light language parameters based on the user's feedback signal on the light language;

[0047] The light language output module is used to output the light language based on the light language parameters in combination with the hardware information and the situational awareness information.

[0048] Due to the adoption of the above technical solution, the beneficial effects achieved by the present invention are as follows:

[0049] 1. In the present invention, the light language parameters include the color, brightness, frequency, and rhythm of light. With its color, brightness, and dynamic changes, light signals can convey emotions more intuitively and create a comfortable or warning atmosphere, thereby improving the efficiency and quality of human-computer interaction. Creating a positive emotional experience through soft blue light or warm yellow light is gentler and more direct than voice, helping to relieve anxiety, promote relaxation, and support the user's mental health. In addition, in noisy environments or when facing hearing-impaired users, light language can provide silent and clear feedback to ensure the effective communication of information and make up for the shortcomings of traditional voice interaction.

[0050] 2. In the present invention, the user's emotional expression information is obtained in advance, and an emotional demand label is generated. Based on the emotional demand label, the light language parameters are generated. By obtaining the user's emotional expression information in advance and generating emotional demand labels, the method can accurately identify the user's current emotional state and generate corresponding light language parameters based on these labels. For example, when it is detected that the user is in an anxious state, the method can generate soft blue light and optimize its transition mode and frequency to help relieve tension.

[0051] 3. In the present invention, based on the light language parameters, combined with hardware information and situational awareness information, the light language is output. Combined with hardware information, this method can maintain a consistent and personalized light language experience on different devices (such as smart lamps, mobile phone screens, wearable devices), ensuring that users can get the same high-quality light language service regardless of the device they use. Through situational awareness information, this method can adjust the light language effect in real time to adapt to different scenes and situational changes, providing the most appropriate atmosphere and support.

[0052] 4. In the present invention, based on the user's feedback signal on the light language, the light language parameters can be continuously adjusted to gradually find the light language mode that best suits the user, providing more accurate and personalized services. This method continuously accumulates data through a user feedback closed loop, optimizes the light language expression, and makes it highly matched with user needs, ensuring the continuous improvement and optimization of the method, providing users with a more intelligent and humanized light language experience, thereby overcoming the limitations of the preset light language expression method in the traditional light language generation method.

[0053] In addition, this method has the ability of adaptive evolution and can continuously learn and optimize in long-term use. It can not only form personalized light language patterns according to the user's emotional response and usage habits, but also make real-time adjustments according to new situational changes, providing highly customized services and breaking through the limitations of static design in traditional light language generation methods. Therefore, this method not only improves the performance of light language, but also opens up new application possibilities in the fields of emotional interaction, intelligent interaction and multimodal experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0055] Figure 1 It is a schematic diagram of the process of the optical language intelligent generation method in one embodiment of the present invention;

[0056] Figure 2 The figure is a logical architecture diagram of the optical language intelligent generation system according to one embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to more clearly illustrate the overall concept of the present invention, a detailed description is given below in an exemplary manner in conjunction with the accompanying drawings.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0059] like Figure 1 As shown, a method for intelligently generating optical language includes:

[0060] S100: Pre-acquire user emotion expression information and generate emotion demand labels.

[0061] In order to accurately capture the user's emotional state, this step needs to collect the user's emotional expression information from multiple dimensions. These information sources include but are not limited to language input, text input, facial expression recognition, and biological data.

[0062] It is understandable that after collecting the user's emotional expression information, a data cleaning step is performed. This includes:

[0063] Remove noise and outliers to ensure data quality;

[0064] Missing data were interpolated or deleted to ensure data integrity.

[0065] Language can record user voice clips for microphone or phone application. Users can express their feelings directly by speaking, providing instant emotional feedback. Voice not only contains literal meaning, but also emotional color, which helps to understand users' emotions more accurately.

[0066] Text is collected from textual content that users enter into applications, social media, or chat interfaces, which may include instant messages, diary entries, etc. Users may be more thoughtful when writing, and the information they provide is often more detailed and specific. Even in the absence of voice input, text input can provide important emotional clues and complement other forms of data.

[0067] Facial expressions can be captured by a camera using a user's facial image or video stream, and computer vision technology can be used to analyze facial feature points, such as changes in eyebrows, eyes, and mouth corners, to identify different emotional states (such as smiling, frowning, wide eyes, etc.). Facial expressions are an intuitive reflection of the user's emotions and can quickly reflect the user's current emotional state. Facial expressions provide a wealth of non-verbal emotional information, which is especially important when the user is unwilling or unable to express it through language. Modern computer vision technology has been able to accurately recognize facial expressions in a variety of complex backgrounds, enhancing the reliability of emotion recognition.

[0068] Biodata is the real-time collection of users’ physiological signals (such as heart rate, breathing rate, body temperature, stress indicators, etc.) using biosensors such as heart rate monitors, breathing rate sensors, and skin conductance sensors. These data can reflect the user’s emotional fluctuations and physical state. Biodata can be used as a means of verifying other emotional expressions to ensure the accuracy of emotion recognition. For example, a faster heart rate may indicate anxiety or excitement, while steady breathing may indicate relaxation.

[0069] Biodata provides objective physiological indicators that are not affected by subjective factors and can more truly reflect the user's emotional state. Biosensors can continuously monitor the user's physical condition, detect emotional changes in a timely manner, and make corresponding adjustments.

[0070] By combining language, text, facial expressions and biological data, the user's emotional state can be fully captured from multiple dimensions. This multimodal data fusion not only improves the accuracy of emotion recognition, but also eliminates the misjudgment that may be caused by a single data source, enhancing the robustness of this method.

[0071] In addition, the various parameters are highly complementary. Language and text provide clear emotional expressions and detailed content descriptions. Facial expressions provide immediate, non-verbal emotional clues. Biological data provides objective physiological evidence to verify other forms of emotional expression.

[0072] The combined use of these parameters can fully cover the user's emotional expressions in different situations. Whether through language, text, facial expressions or physiological reactions, the most complete emotional information can be captured, thereby generating more accurate emotional demand labels.

[0073] The method further comprises:

[0074] S200: Generate light language parameters based on the emotional demand tag, where the light language parameters include light color, brightness, frequency, and rhythm.

[0075] In this step, first, emotional demand labels (such as "relaxation", "anxiety", "excitement", etc.) are generated based on the pre-acquired user emotional expression information. These labels provide clear guidance for the subsequent generation of light language parameters.

[0076] Light language parameters mainly include:

[0077] Colors: Choose colors that are appropriate to express your emotions, such as warm colors for relaxation and cool colors for relieving anxiety;

[0078] Brightness, set the appropriate brightness level, ranging from soft low brightness to eye-catching high brightness;

[0079] Frequency, set the flashing speed of the light signal, such as fast flash, slow flash or no flash;

[0080] Rhythm, which defines the time series characteristics of light signals, such as periodic alternation of light and dark or non-periodic random fluctuations, enhances the user experience.

[0081] Different colors can trigger different emotional responses. For example, warm colors (such as red, orange, and yellow) are usually associated with warmth, vitality, and comfort; cool colors (such as blue and green) are associated with calmness, relaxation, and serenity. Choosing the most appropriate color based on the user's emotional need label can effectively adjust the user's psychological state. Color has a significant impact on people's emotions. By adjusting the color, the environment atmosphere can be quickly changed to help users better manage their emotions.

[0082] Brightness directly affects the user's visual comfort. Too high or too low brightness can cause discomfort, while appropriate brightness can provide a comfortable visual experience. Brightness changes can enhance or weaken emotional effects. For example, low-brightness soft light helps to relax, while high-brightness light can improve concentration and alertness. Brightness changes can further enhance the emotional expression of color and respond to users' emotional needs more accurately.

[0083] Frequency determines how fast the light signal flashes. Fast flashes can convey excitement or warning messages, while slow fades can create a calm and relaxing atmosphere. Different frequencies can affect the user's physiological response. By adjusting the frequency, you can guide the user into a specific emotional state, such as relaxation, concentration or excitement.

[0084] Rhythm defines the time series characteristics of the light signal, such as periodic alternation of light and dark or non-periodic random fluctuations. This change can simulate natural phenomena (such as sunrise and sunset) or musical rhythms to enhance emotional resonance. Rhythm can be synchronized with the user's physiological indicators such as heart rate and breathing rate to help users better manage their emotions. By fine-tuning the rhythm, it can be personalized according to the user's real-time emotional changes to provide the best light language expression.

[0085] By combining color, brightness, frequency and rhythm, the user's emotional demand labels can be fully expressed from multiple dimensions. This multi-dimensional expression not only improves the accuracy of emotional transmission, but also creates a richer and more delicate light language experience. These parameters not only affect the user's emotional perception, but are also closely related to physiological reactions. This synergy of emotion and physiology enables light language to not only convey emotions, but also actually affect the user's physical state.

[0086] S300: Outputting the light language based on the light language parameters in combination with the hardware information and the situational awareness information.

[0087] In order to ensure the consistency and stability of light language on different hardware platforms, it is necessary to make appropriate adjustments based on the hardware characteristics. At the same time, in order to achieve the best expression effect of light language, it is necessary to perceive and adapt to complex dynamic environmental changes in real time.

[0088] Among them, it automatically identifies the type of connected device (such as smart lamps, mobile phone screens, wearable devices) and obtains its key performance indicators (such as processor speed, memory capacity, light source intensity, display resolution, etc.). According to the hardware performance, it adjusts the dynamic change frequency and complexity of the optical signal to ensure smooth operation without exceeding the device processing capacity.

[0089] In addition, multiple sensors (such as ambient light sensors, temperature and humidity sensors) are integrated to collect environmental data in real time. Deep learning models are used to analyze multimodal data to identify user activity patterns and specific environmental contexts (such as day / night, work / leisure).

[0090] In this step, in order to ensure that the light language can accurately convey emotions and support user needs, the light language output needs to be executed efficiently. According to the communication protocol supported by the device (such as Wi-Fi, Bluetooth, Zigbee, etc.), the most suitable transmission method is selected to send the optimized light language parameters to the target hardware. The parameters are converted into control instructions that the device can understand, such as PWM (pulse width modulation) signals for adjusting LED brightness, or API calls for setting screen colors.

[0091] In this step, by adapting to different hardware characteristics, a consistent and personalized light language experience can be maintained on multiple devices (such as smart lamps, mobile phone screens, and wearable devices), ensuring that users can get the same high-quality service regardless of the device they use. Parameters are adjusted according to the characteristics of each device, so that the light language effect meets the performance limitations of the device and maximizes the visual effect and emotional transmission.

[0092] By monitoring the ambient light, temperature and humidity in real time, the light language effect can be dynamically adjusted to adapt to different scenes and situational changes, providing the most suitable atmosphere and support. The brightness and color temperature are automatically adjusted according to the environmental conditions to ensure a comfortable visual experience at different times and places. For example, the brightness is reduced at night to avoid glare, and cool lighting is used in high temperature environments to create a cool feeling.

[0093] S400: Based on the user's feedback signal on the light language, adjust the light language parameters to output a new light language.

[0094] In order to ensure that the system's optical language can be continuously optimized and better meet user needs, this step requires real-time collection of user feedback signals.

[0095] It should be noted that the feedback signal includes user emotional expression information, that is, monitoring the user's emotional state (such as heart rate, breathing rate, etc.), and also includes the user's satisfaction evaluation, that is, collecting the user's rating of the light language effect through an application or other interactive interface.

[0096] Integrate feedback signals from different sources to form a comprehensive user feedback feature vector. Apply natural language processing (NLP) technology and machine learning models to classify the user's specific opinions into sentiments, such as "positive", "negative" or "neutral". Combine physiological data (such as heart rate variability HRV, skin conductance, etc.) to verify the user's emotional state and ensure the accuracy of the feedback signal.

[0097] According to the user's feedback signal, the system adopts timely adjustment strategy to adjust the light language parameters:

[0098] For obvious negative feedback (such as users clearly indicating that the light is too bright or the color is uncomfortable), immediately adjust the corresponding light language parameters, such as reducing the brightness or changing the color.

[0099] For more vague feedback (such as users feeling some discomfort but not clearly stating the reason), gradually fine-tune the parameters, observe user reactions, and find the most appropriate settings.

[0100] The adjusted light language parameters are used to generate new light language effects to ensure that they better meet the actual needs of users.

[0101] In order to ensure that the adjusted light language effect can be continuously optimized, a user feedback closed-loop mechanism is established in this step. The user's satisfaction rating and other forms of feedback on the new light language effect are collected to further optimize the light language expression mode. If the user is satisfied, the current setting is maintained; if there is any discomfort, the system will continue to adjust until the best effect is achieved.

[0102] As a preferred implementation of the present invention, the user's emotional expression information is obtained in advance and an emotional demand label is generated, specifically:

[0103] The user's emotional expression information includes at least one of language, text, facial expression, and biological data;

[0104] Extracting feature parameters corresponding to the user's emotional expression information, generating a feature vector including at least one of the feature parameters, and generating an emotional demand label based on a deep learning model;

[0105] The emotional need labels include relaxation, anxiety, or excitement.

[0106] In this implementation, for each type of user emotion expression information, it is necessary to extract corresponding feature parameters to form a feature vector as a basis for subsequent processing.

[0107] The extracted feature parameters are integrated into a multi-dimensional feature vector for subsequent emotion recognition and emotion demand label generation, such as [language feature vector, text feature vector, facial expression feature vector, biological data feature vector].

[0108] The feature vector is input into the trained deep learning model to predict that the user is currently in a "relaxed" emotional state, so as to generate an "relaxed" emotional demand label and pass it to the light language generation module. Of course, in this embodiment, it is also possible to predict that the user is currently in an "anxious" or "excited" emotional state.

[0109] Relaxation refers to a state of mind in which an individual is calm, comfortable, and stress-free. In this state, the body's tension is reduced, the heart rate slows down, the breathing is steady, and the muscles are relaxed. It is more common in scenarios such as preparation for bed, meditation, and leisure time.

[0110] Anxiety is a feeling of worry about potential threats or uncertainties, often accompanied by tension, uneasiness, and increased alertness. Physiologically, it may manifest as increased heart rate, rapid breathing, sweating, etc., and is more common in situations such as work pressure, before exams, and before public speeches.

[0111] Excitement is a positive emotional state, usually accompanied by pleasure, vitality and high alertness. Physiologically, it may manifest as increased heart rate, accelerated breathing, muscle tension, etc. It is more common in celebrations, before and after exercise, and during creative work.

[0112] Of course, other emotional demand labels can also be further defined in detail, and this embodiment does not limit this.

[0113] As a preferred embodiment of this implementation, the feature parameters corresponding to the user's emotional expression information are extracted, specifically:

[0114] The biological data includes heart rate data or respiration data, and the heart rate variability corresponding to the heart rate data and the respiration frequency corresponding to the respiration data are calculated to generate characteristic parameters corresponding to the biological data;

[0115] Extracting sound features corresponding to the language, the sound features including fundamental frequency, energy or zero-crossing rate, to generate feature parameters corresponding to the language;

[0116] Applying natural language processing technology to perform part-of-speech tagging or sentiment classification on the text to generate feature parameters corresponding to the text;

[0117] Based on a computer vision algorithm, facial features that can map emotion categories are extracted to generate feature parameters corresponding to the facial expressions.

[0118] Among them, heart rate variability refers to the change in the heartbeat interval time, which reflects the activity state of the autonomic nervous system. Heart rate variability is obtained by analyzing the continuous heartbeat interval time (RR interval), calculating the standard deviation (SDNN), the root mean square difference of adjacent RR intervals (RMSSD), low frequency / high frequency power ratio and other indicators. Heart rate variability is closely related to emotions. High heart rate variability usually indicates relaxation and calmness, while low heart rate variability may indicate anxiety or stress.

[0119] Respiratory rate refers to the number of breaths per unit time, which reflects the user's physiological state. The sensor monitors the user's breathing cycle in real time and calculates the number of breaths per minute. Respiratory rate is directly related to emotional state. For example, the breathing rate usually speeds up when you are anxious, and tends to be stable when you are relaxed.

[0120] Specifically, the heart rate monitor detects that the user's heart rate is 75 beats / minute, the standard deviation of the RR interval is 50 milliseconds, and the RMSSD is 30 milliseconds. The respiratory rate sensor detects that the user's respiratory rate is 16 times / minute. Generate a heart rate variability feature vector [SDNN = 50ms, RMSSD = 30ms] and a respiratory rate feature vector [respiratory rate = 16 times / minute].

[0121] The fundamental frequency is the main frequency component of the speech signal, reflecting the frequency of the vocal cord vibration. Automatic speech recognition (ASR) technology is used to extract the fundamental frequency in the speech signal. The fundamental frequency can reflect the speaker's emotional state. For example, the fundamental frequency may increase when anxious and decrease when relaxed.

[0122] Energy is the amplitude of a speech signal, reflecting the strength of the sound. Calculate the short-term energy of a speech signal, that is, the average amplitude in each small time period. High energy usually indicates excitement or anger, while low energy may indicate fatigue or relaxation.

[0123] Zero Crossing Rate (ZCR) is the frequency at which a speech signal crosses zero, reflecting the roughness of the sound. Count the number of times a speech signal crosses zero. A high ZCR usually indicates tension or excitement, while a low ZCR may indicate calmness or relaxation.

[0124] Specifically, the user says: "I feel a little tired today." The speech processing module extracts the fundamental frequency = 120 Hz, energy = 0.8, and zero crossing rate = 5 times / second, and generates a sound feature vector [fundamental frequency = 120 Hz, energy = 0.8, and zero crossing rate = 5 times / second].

[0125] In the text feature parameter extraction, natural language processing (NLP) is used to identify words and their grammatical roles in the text, such as nouns, verbs, adjectives, etc., and pre-trained language models (such as BERT) are used for part-of-speech tagging. The emotional color of specific words can help understand the user's emotional state, for example, words such as "tired" and "relaxed".

[0126] Sentiment classification refers to classifying text into positive, negative, or neutral sentiment categories. Use a pre-trained sentiment classification model (such as BERT) to classify text sentiment. Sentiment classification can directly reflect the user's emotional tendencies, such as "happy", "sad", "angry", etc.

[0127] Specifically, a user writes on a mobile phone application: "I need to relax for a while." The text processing module annotates the words: "need" (verb), "relax" (verb), "a while" (adverb). The sentiment classification model predicts that the text is "positive", indicating that the user wants to relax. The text feature parameters [sentiment vocabulary = "relax", sentiment classification = "positive"] are generated.

[0128] In the extraction of facial expression feature parameters, key points of the face, such as eyebrows, eyes, corners of the mouth, etc., are identified and tracked. Convolutional Neural Network (CNN) is used to detect facial feature points. Different facial feature changes can be mapped to different emotion categories, such as smiling, frowning, wide eyes, etc.

[0129] Capture dynamic changes in facial expressions, such as blinking frequency, mouth opening and closing, etc. Combined with time series analysis, capture the temporal change pattern of facial expressions. Dynamic changes can provide richer non-verbal emotional information, such as fast blinking may indicate tension, slow blinking may indicate relaxation.

[0130] Specifically, the camera captures the user's slightly tired expression. The computer vision module detects that the eyebrows are slightly drooping, the eyes are half closed, and the corners of the mouth are relaxed. It detects that the blinking frequency is slow and the mouth is occasionally open. Generate a facial expression feature vector [facial feature point = "tired expression", dynamic change = "slightly frowned, slow blinking frequency"].

[0131] It should be noted that after extracting the feature parameters corresponding to biological data, language, text, and facial expressions, respectively, the extracted feature parameters are integrated into a multidimensional feature vector to make a comprehensive judgment on the emotional demand label.

[0132] In this embodiment, by finely extracting data from different sources, the user's emotional state can be captured more comprehensively and the accuracy of emotion recognition can be improved. Combining multiple feature parameters, more personalized emotional demand labels can be generated to meet the needs of different users. Based on the deep learning model, the system has adaptive evolution capabilities and can continuously optimize its emotion recognition capabilities during long-term use. Through accurate emotional demand labels, the system can provide a light language experience that is more in line with user needs, significantly improving the quality of user experience and human-computer interaction.

[0133] As a preferred implementation of the present invention, based on the emotional demand label, light language parameters are generated, specifically:

[0134] Establishing a mapping relationship between the emotional demand label and the light language parameter, and setting initial light language parameters for the emotional demand label one by one;

[0135] The historical data is called up to analyze the user's usage habits, and the mapping relationship is adjusted according to the user's usage habits.

[0136] For each emotional demand label, a set of initial light language parameters needs to be pre-set to ensure a reasonable light language experience when used for the first time.

[0137] Relaxation: soft warm yellow light, moderately reduced brightness, no obvious flicker, slow gradient;

[0138] Anxiety: soft blue light, moderate brightness adjustment, fast gradient, smooth changes;

[0139] Excitement: bright and colorful lights, higher brightness, high-frequency flashing, and rapid changes.

[0140] In order to achieve more personalized services, it is necessary to call on historical data, analyze user usage habits, build user portraits, and dynamically adjust the mapping relationship based on the user portraits.

[0141] Based on historical data, we can identify users’ long-term preferences and habits. For example, some users may prefer to use soft blue light at night, while others prefer warm yellow light. Based on the accumulated data, we can gradually learn users’ unique emotional characteristics and preferences, form an optimization plan for individuals, and provide a highly customized light language experience.

[0142] In addition, through precise mapping relationship adjustment, the system can provide a light language experience that is more in line with user needs, significantly improving the quality of user experience and human-computer interaction. The continuous optimization of the system and the improvement of service quality have enhanced user loyalty and satisfaction.

[0143] As a preferred implementation of the present invention, based on the light language parameters, combined with hardware information and situational awareness information, the light language is output, specifically:

[0144] According to a preset hardware fingerprint library, the connected hardware device is identified, the corresponding hardware information is obtained, and the optical language parameters are adjusted according to a standardized optical language parameter library covering a variety of hardware information;

[0145] By using a variety of sensors, environmental data is collected in real time, combined with the emotional demand tags, to form situational awareness information, and the light language parameters are adjusted according to the situational awareness information;

[0146] The light language parameters are transmitted to a connected hardware device so as to output the light language through the hardware device.

[0147] The hardware fingerprint library is a database containing characteristic information of various hardware devices, which is used to uniquely identify and describe each device, including device type (such as smart lamps, mobile phone screens, wearable devices), manufacturer, model, performance indicators (such as processor speed, memory capacity, light source intensity, display resolution, etc.).

[0148] When a new device is connected, the system automatically obtains the device's unique identifier (such as MAC address, serial number) through a network protocol (such as Wi-Fi, Bluetooth, Zigbee, etc.) and searches for a match in the hardware fingerprint library. Once the identification is successful, the system extracts all relevant information about the device from the hardware fingerprint library, including supported color modes (RGB, HSL, etc.), maximum brightness output capability, dynamic effect support, etc.

[0149] The standardized light language parameter library is a standardized parameter library covering a variety of hardware information, ensuring the consistency and optimal performance of light language on different devices, including color mode (RGB, HSL), brightness range (0-100%), frequency range (0-100Hz), rhythm mode (periodic, non-periodic), etc.

[0150] Based on the hardware information obtained, the system selects the most appropriate parameter combination from the standardized light language parameter library to ensure the best performance of the light language on different devices.

[0151] Integrate multiple sensors (such as ambient light sensor, temperature and humidity sensor) to collect environmental data in real time, including ambient light intensity, temperature, humidity, etc.

[0152] Based on deep learning algorithms, combined with the user's emotional demand labels and environmental data, the current usage scenario (such as sleep, work, entertainment, etc.) is judged in real time, and then the light language parameters are optimized.

[0153] This implementation can provide a consistent and personalized light language experience on multiple devices, ensuring the continuity and stability of the user experience, and can perceive and adapt to complex dynamic environmental changes in real time, provide the most appropriate light language expression, and enhance user comfort and emotional resonance.

[0154] In addition, by fine-tuning the light language parameters based on the user's physiological response and environmental changes, users can better manage their emotions and achieve psychological balance and stability. Moreover, through precise situational awareness and parameter adjustment, the system can provide a light language experience that better meets user needs, significantly improving the quality of user experience and human-computer interaction.

[0155] As an example of this implementation, the optical language parameters are adjusted according to a standardized optical language parameter library covering a variety of hardware information, specifically:

[0156] The hardware information includes color modes, light source intensity, and performance parameters supported by the hardware device;

[0157] Adjust the color parameters in the light language parameters according to the color mode supported by the hardware device;

[0158] Adjust the brightness parameter in the light language parameters according to the light intensity of the hardware device;

[0159] According to the performance parameters of the hardware equipment, adjust the frequency parameters and rhythm parameters in the light language parameters.

[0160] Among them, color mode refers to the color representation method supported by the device, such as RGB (red, green, blue), HSL (hue, saturation, brightness), HSV (hue, saturation, brightness), etc.

[0161] Light intensity refers to the maximum light output capability of a device, usually expressed in lumens (lm) or as a percentage (0-100%).

[0162] Performance parameters refer to the device's ability to handle dynamic effects, including maximum refresh rate, supported dimming technology (such as PWM pulse width modulation), maximum frequency range, etc.

[0163] If the device supports RGB mode, you can directly use RGB values ​​to set color parameters. For example, soft warm yellow light can be set to RGB = [255, 204, 102]. If the device supports HSL or HSV mode, you need to convert the RGB value to the corresponding HSL or HSV value. For example, soft warm yellow light can be converted to HSL = [48°, 76%, 70%] or HSV = [48°, 60%, 100%].

[0164] Set the appropriate brightness level based on the device's maximum brightness output capability. For example, if the device's maximum brightness is 800 lumens, you can adjust it between 0 and 800 lumens. Combined with data from the ambient light sensor, dynamically adjust the brightness to maintain an appropriate contrast ratio. For example, increase the brightness in a bright environment and reduce it in a dim environment.

[0165] Specifically, the device has a maximum brightness of 800 lumens, and sets a soft warm yellow light with a brightness of 60%, or 480 lumens. When it detects that the light in the room is dim, the brightness is appropriately increased to 70%, or 560 lumens.

[0166] According to the maximum refresh rate of the device, set the appropriate frequency parameters. For example, if the maximum refresh rate of the device is 100Hz, you can select a frequency between 0 and 100Hz. If the device supports PWM dimming, you can use the PWM signal to achieve smooth brightness changes. For example, set the frequency to 30Hz to ensure a smooth transition effect. According to the performance parameters of the device, adjust the complexity of the dynamic changes of the light signal to ensure smooth operation without exceeding the processing capacity of the device.

[0167] Specifically, the maximum refresh rate of the device is 100Hz, and the frequency is set to 30Hz to ensure a smooth transition effect; for high-performance devices, set a fast-changing rhythm and a frequency of 60Hz to enhance dynamics and vitality; for low-performance devices, simplify the dynamic effects and set the frequency to 20Hz to ensure smooth operation.

[0168] As another embodiment of this implementation, environmental data is collected in real time through multiple sensors, combined with the emotional demand label to form situational awareness information, and the light language parameters are adjusted according to the situational awareness information, specifically:

[0169] The environmental data includes ambient light, temperature and humidity.

[0170] According to the environmental data, combined with the emotional demand label, situational awareness information is formed,

[0171] According to the ambient light, adjust the brightness of the light language parameter corresponding to the emotional demand label;

[0172] According to the ambient temperature and humidity, the color in the light language parameter corresponding to the emotional demand label is adjusted.

[0173] Among them, the ambient light sensor includes a photodiode, a photoresistor, etc., which is used to measure the ambient light intensity (lux), color temperature (K), etc.

[0174] Temperature and humidity sensors include DHT22, SHT30, etc., which are used to measure data including temperature (℃ or °F) and relative humidity (%RH).

[0175] Based on deep learning algorithms, combined with the user's environmental data, the current usage scenario (such as sleep, work, entertainment, etc.) is judged in real time according to the user's emotional demand labels (such as relaxation, anxiety, excitement).

[0176] Specifically, the user is in a state of relaxation, the light in the room is dim, the temperature is 28°C, the humidity is 75%, and the heart rate and breathing rate tend to be stable. The system determines that the current scene is a relaxation scene before bed;

[0177] The user feels anxious, the office is brightly lit, the temperature is 24°C, the humidity is 50%, the heart rate and breathing rate increase. The system determines that the current scene is anxiety at work;

[0178] The user is at a family gathering or party, the ambient light is moderate, the temperature is 26℃, the humidity is 60%, the heart rate is accelerated, and the breathing rate is normal. The system determines that the current scene is exciting during entertainment.

[0179] According to the ambient light, the brightness in the light language parameter corresponding to the emotional demand tag is adjusted. In combination with the data of the ambient light sensor, the brightness is dynamically adjusted to maintain a suitable contrast. For example, the brightness is increased in a bright environment and decreased in a dim environment.

[0180] Specifically, if the ambient light is strong (>500lux), increase the brightness appropriately to ensure clear vision but not too glaring. If the ambient light is weak (<100lux), reduce the brightness appropriately to avoid excessive brightness that irritates the eyes and create a comfortable atmosphere.

[0181] According to the ambient temperature and humidity, adjust the color in the light language parameter corresponding to the emotional demand label. Adjust the color according to the ambient temperature to help users better adapt to environmental changes. Use cool light (such as blue or blue-green) in a high temperature environment to create a cool feeling. Use warm light (such as yellow or orange) in a low temperature environment to create a warm feeling.

[0182] Fine-tune the color according to the ambient humidity to improve comfort. Use soft, cool lighting in high-humidity environments to reduce the feeling of stuffiness. Use slightly warmer tones in low-humidity environments to increase comfort.

[0183] This embodiment can provide a consistent and personalized light language experience under different environmental conditions, ensuring the continuity and stability of the user experience. In addition, it can perceive and adapt to complex dynamic environmental changes in real time, provide the most appropriate light language expression, and enhance the user's comfort and emotional resonance.

[0184] In addition, through precise situational awareness and parameter adjustment, the system can provide a light language experience that is more in line with user needs, significantly improving the quality of user experience and human-computer interaction.

[0185] As a preferred implementation of the present invention, based on the user's feedback signal on the light language, the light language parameters are adjusted to output a new light language, specifically:

[0186] The feedback signal includes user emotion expression information,

[0187] Collecting the user's emotional expression information in real time during the light language output process, and adjusting the light language parameters according to the trend change of the user's emotional expression information;

[0188] and / or, the feedback signal includes a user's satisfaction evaluation,

[0189] The light language parameters are adjusted according to the user's satisfaction evaluation.

[0190] The sources of information about user emotional expressions include: language input, recording user voice clips through a microphone or phone application;

[0191] Text input, collected from text content entered by users in applications, social media or chat interfaces;

[0192] Facial expression recognition, using a camera to capture a user's facial image or video stream;

[0193] Biological data uses heart rate monitors, respiratory rate sensors, skin conductance sensors and other devices to collect the user's physiological signals in real time.

[0194] The user's emotional expression information is collected in real time through the integration of multiple sensors (such as ambient light sensors, temperature and humidity sensors, heart rate and breathing monitors). Combined with information from multiple sources such as language, text, facial expressions and biological data, a comprehensive emotional expression feature vector is formed. The time series characteristics of the user's emotional expression information are analyzed, such as the fluctuation pattern of heart rate, the change trend of breathing frequency, etc. Natural language processing (NLP) technology and machine learning models are used to classify the user's specific opinions into sentiments, such as "positive", "negative" or "neutral".

[0195] In addition, users' ratings of the light language effects (e.g., 1 to 5 stars) are collected through applications or other interactive interfaces. A text input box or selection menu is provided to allow users to describe their feelings and suggestions in detail, such as "the light is too bright" or "the color is not soft enough." Natural language processing (NLP) technology and machine learning models are used to classify the sentiment of users' specific opinions, such as "positive," "negative," or "neutral."

[0196] For obvious negative feedback (such as users clearly indicating that the light is too bright or the color is uncomfortable), immediately adjust the corresponding light language parameters to quickly improve the user experience. For more vague feedback (such as users feeling a little uncomfortable but not clearly stating the reason), gradually fine-tune the parameters, observe the user's reaction, and find the most appropriate settings. This implementation uses the user's emotional expression information as a reward signal in the reinforcement learning framework to quantify the user's emotional changes. By constantly trying different combinations of light language parameters and evaluating based on user feedback, the best light language expression mode is gradually found.

[0197] This implementation can quickly correct negative user feedback and improve user satisfaction and willingness to use. Fine-tune light language parameters based on user physiological responses and environmental changes to help users better manage their emotions and achieve psychological balance and stability. Through precise feedback signal adjustment, it can provide a light language experience that better meets user needs, significantly improving the quality of user experience and human-computer interaction.

[0198] As a preferred embodiment of this implementation, the mapping relationship between the emotion demand label and the light language parameter is iterated according to the feedback signal.

[0199] The system accumulates user feedback data over a long period of time, including information such as emotional responses, usage habits, and environmental changes.

[0200] In this embodiment, user feedback data is accumulated over a long period of time, including information on emotional expression, usage habits, and environmental changes. Based on the accumulated data, the system gradually learns the user's unique emotional characteristics and preferences, forms an optimization plan for the individual, and provides a highly customized light language experience.

[0201] Specifically, deep learning models (such as neural networks, decision trees, etc.) are used to train the accumulated historical data to optimize the mapping relationship between emotional demand labels and light language parameters, so as to have adaptive evolution capabilities and continuously optimize its light language expression mode during long-term use.

[0202] This embodiment can provide a highly personalized light language experience based on the user's long-term usage habits and real-time feedback to meet the needs of different users. Based on the deep learning model and reinforcement learning framework, the system has adaptive evolution capabilities and can continuously optimize its light language expression mode in long-term use.

[0203] In addition, through precise mapping relationship adjustment, it can provide a light language experience that is more in line with user needs, significantly improving the quality of user experience and human-computer interaction. The continuous optimization of the system and the improvement of service quality have enhanced user loyalty and satisfaction.

[0204] like Figure 2 As shown, the present invention again provides a light language intelligent generation system, including:

[0205] Emotional demand generation module, used to obtain user emotional expression information and generate emotional demand labels;

[0206] A light language parameter generating module, used to generate light language parameters based on the emotional demand label, and adjust the light language parameters based on the user's feedback signal on the light language;

[0207] The light language output module is used to output the light language based on the light language parameters in combination with the hardware information and the situational awareness information.

[0208] The optical language intelligent generation system can achieve any technical effect of the above-mentioned optical language intelligent generation method, which will not be repeated here.

[0209] Anything not described in the present invention can be achieved by adopting or drawing on existing technologies.

[0210] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0211] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for intelligently generating optical language, characterized in that: Pre-acquiring user emotion expression information and generating an emotion demand label, the method further includes: Based on the emotional demand tag, generating light language parameters, the light language parameters including light color, brightness, frequency, and rhythm; Based on the light language parameters, combined with the hardware information and the situational awareness information, output the light language; Based on the user's feedback signal on the light language, the light language parameters are adjusted to output a new light language.

2. The method for intelligently generating optical language according to claim 1, characterized in that: Obtain user emotional expression information in advance and generate emotional demand labels, specifically: The user's emotional expression information includes at least one of language, text, facial expression, and biological data; Extracting feature parameters corresponding to the user's emotional expression information, generating a feature vector including at least one of the feature parameters, and generating an emotional demand label based on a deep learning model; The emotional need labels include relaxation, anxiety, or excitement.

3. The method for intelligently generating optical language according to claim 2, characterized in that: Extract the characteristic parameters corresponding to the user's emotional expression information, specifically: The biological data includes heart rate data or respiration data, and the heart rate variability corresponding to the heart rate data and the respiration frequency corresponding to the respiration data are calculated to generate characteristic parameters corresponding to the biological data; Extracting sound features corresponding to the language, the sound features including fundamental frequency, energy or zero-crossing rate, to generate feature parameters corresponding to the language; Applying natural language processing technology to perform part-of-speech tagging or sentiment classification on the text to generate feature parameters corresponding to the text; Based on a computer vision algorithm, facial features that can map emotion categories are extracted to generate feature parameters corresponding to the facial expressions.

4. The method for intelligently generating optical language according to claim 1, characterized in that: Based on the emotional demand label, light language parameters are generated, specifically: Establishing a mapping relationship between the emotional demand label and the light language parameter, and setting initial light language parameters for the emotional demand label one by one; The historical data is called up to analyze the user's usage habits, and the mapping relationship is adjusted according to the user's usage habits.

5. The method for intelligently generating optical language according to claim 1, characterized in that: Based on the light language parameters, combined with the hardware information and the situational awareness information, the light language is output, specifically: According to a preset hardware fingerprint library, the connected hardware device is identified, the corresponding hardware information is obtained, and the optical language parameters are adjusted according to a standardized optical language parameter library covering a variety of hardware information; By using a variety of sensors, environmental data is collected in real time, combined with the emotional demand tags, to form situational awareness information, and the light language parameters are adjusted according to the situational awareness information; The light language parameters are transmitted to a connected hardware device so as to output the light language through the hardware device.

6. The method for intelligently generating optical language according to claim 5, characterized in that: According to a standardized optical language parameter library covering a variety of hardware information, the optical language parameters are adjusted, specifically: The hardware information includes color modes, light source intensity, and performance parameters supported by the hardware device; Adjust the color parameters in the light language parameters according to the color mode supported by the hardware device; Adjust the brightness parameter in the light language parameters according to the light intensity of the hardware device; According to the performance parameters of the hardware equipment, adjust the frequency parameters and rhythm parameters in the light language parameters.

7. The method for intelligently generating optical language according to claim 5, characterized in that: Through a variety of sensors, environmental data is collected in real time, combined with the emotional demand tags, to form situational awareness information, and the light language parameters are adjusted according to the situational awareness information, specifically: The environmental data includes ambient light, temperature and humidity. According to the environmental data, combined with the emotional demand label, situational awareness information is formed, According to the ambient light, adjust the brightness of the light language parameter corresponding to the emotional demand label; According to the ambient temperature and humidity, the color in the light language parameter corresponding to the emotional demand label is adjusted.

8. The method for intelligently generating optical language according to claim 1, characterized in that: Based on the user's feedback signal on the light language, the light language parameters are adjusted to output a new light language, specifically: The feedback signal includes user emotion expression information, Collecting the user's emotional expression information in real time during the light language output process, and adjusting the light language parameters according to the trend change of the user's emotional expression information; and / or, the feedback signal includes a user's satisfaction evaluation, The light language parameters are adjusted according to the user's satisfaction evaluation.

9. The method for intelligently generating optical language according to claim 8, characterized in that: According to the feedback signal, the mapping relationship between the emotion demand label and the light language parameter is iterated.

10. An optical language intelligent generation system, characterized in that: include, Emotional demand generation module, used to obtain user emotional expression information and generate emotional demand labels; A light language parameter generating module, used to generate light language parameters based on the emotional demand label, and adjust the light language parameters based on the user's feedback signal on the light language; The light language output module is used to output the light language based on the light language parameters in combination with the hardware information and the situational awareness information.

Images