A music adaptive adjustment method and device

Abstract

This application discloses a music adaptive adjustment method and device, belonging to the field of artificial intelligence technology. It addresses the problem that when background music is mismatched with learning activities, it can interfere with the learning process and reduce learning efficiency. The method includes: performing task matching and analysis based on learning scene information obtained from smart glasses to obtain learning task information; wherein the learning task information includes at least one of the following: learning task type, learning task duration, and learning task difficulty level; analyzing and detecting real-time learning behavior information obtained from the smart glasses to obtain a real-time learning state; and matching corresponding background music from a music database based on the learning task information and / or the real-time learning state to perform adaptive background music adjustment.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a music adaptive adjustment method and device. Background Technology

[0002] In modern learning environments, background music is widely used to optimize the learning experience. Appropriate background music can effectively improve learners' concentration by regulating the activity of brain nerves, while also accelerating information processing efficiency, making the learning process more efficient.

[0003] However, most existing music playback systems use fixed playlists or recommendations based on user preferences. This can lead to a mismatch between music and learning activities. For example, playing overly energetic music during deep reading that requires high concentration, or playing overly soothing music when the user is fatigued, can interfere with the learning process and reduce learning efficiency. Summary of the Invention

[0004] This application provides a music adaptive adjustment method and device to solve the following technical problem: when the background music does not match the learning activity, the background music may interfere with the learning process and reduce learning efficiency.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] This application provides a method for adaptive music adjustment. It includes: performing task matching and analysis based on learning scenario information acquired by smart glasses to obtain learning task information; wherein the learning task information includes at least one of the following: learning task type, learning task duration, and learning task difficulty level; analyzing and detecting real-time learning behavior information acquired by the smart glasses to obtain a real-time learning state; and matching corresponding background music from a music database based on the learning task information and / or the real-time learning state to perform adaptive background music adjustment.

[0007] In one implementation of this application, task matching and analysis are performed based on the learning scenario information obtained by the smart glasses to obtain learning task information. Specifically, when determining the learning task type: image recognition processing is performed on the image data in the learning scenario information to extract visual feature information; wherein, the visual feature information includes at least one of the following: learning tools, learning materials, user learning posture, and text data generated by the user during learning; text recognition is performed on the text data in the learning scenario information using natural language processing technology to extract text feature information; the visual feature information and text feature information are fused to obtain key learning information; the key learning information is matched in a pre-set learning task database to determine the learning task type; wherein, the learning task type includes at least one of the following: deep reading, language learning, mathematical problem solving, and writing.

[0008] In one implementation of this application, task matching and analysis are performed based on the learning scenario information obtained by the smart glasses to obtain learning task information. Specifically, when determining the duration of the learning task: a first reference learning task duration is determined based on the amount of text content in the learning materials; a second reference learning task duration preset by the user is obtained; historical learning data that meets the similarity criteria with the learning key information is identified in the historical database; the learning key information, the first reference learning task duration, the second reference learning task duration, and the historical learning data are input into a preset duration prediction model to output the learning task duration corresponding to the current learning task.

[0009] In one implementation of this application, task matching and analysis are performed based on the learning scenario information obtained by the smart glasses to obtain learning task information. Specifically, when determining the difficulty level of the learning task: the complexity level of the learning task is determined according to the type of text content in the learning materials; historical learning data corresponding to the complexity level of the learning task is determined in the historical database; based on the completion result data corresponding to the historical learning data, a learning profile and learning curve corresponding to the user are constructed under the complexity level of the learning task, so as to determine the difficulty level of the learning task according to the user profile and learning curve; wherein, the completion result data includes at least the task completion rate, the task error rate, and the thinking time.

[0010] In one implementation of this application, the real-time learning behavior information acquired by the smart glasses is analyzed and detected to obtain the real-time learning state. Specifically, this includes: acquiring real-time learning behavior information sent by different information acquisition devices on the smart glasses; wherein the real-time learning behavior information includes at least one of the following: eye-tracking information, head posture information, facial expression information, and physiological feature parameters; comparing different types of real-time learning behavior information with corresponding state thresholds, and determining the state coefficient value corresponding to each type of real-time learning behavior information based on the comparison difference; and obtaining the real-time learning state based on preset weight allocation parameters, real-time learning behavior information, and state coefficient values; wherein the real-time learning state includes a distracted state and a fatigued state.

[0011] In one implementation of this application, the real-time learning state is obtained based on preset weight allocation parameters, real-time learning behavior information, and state coefficient values. Specifically, this includes: selecting the type of real-time learning behavior information required for distraction detection from the real-time learning behavior information to construct a distraction detection information set; weighting each piece of real-time learning behavior information in the distraction detection information set based on the weight allocation parameters and state coefficient values ​​corresponding to the distraction detection information set to obtain the user's distraction state; selecting the type of real-time learning behavior information required for fatigue detection from the real-time learning behavior information to construct a fatigue detection information set; weighting each piece of real-time learning behavior information in the fatigue detection information set based on the weight allocation parameters and state coefficient values ​​corresponding to the fatigue detection information set to obtain the user's fatigue state.

[0012] In one implementation of this application, based on learning task information and / or real-time learning status, corresponding background music is matched in a music database, and adaptive adjustment of the background music is performed. Specifically, this includes: determining a first music set in the music database based on the learning task type; constructing a music characteristic trend curve based on the learning task duration and difficulty level, and determining a second music set in the music database based on the music characteristic trend curve; wherein the music characteristic trend curve is divided into multiple playback stages based on changes in music characteristics; in the multiple playback stages, determining a learning background music list based on the intersection of the first and second music sets, and playing it; determining a third music set in the music dataset based on the real-time learning status; and adjusting the learning background music list based on the third music set.

[0013] In one implementation of this application, before determining and playing the learning background music list, the method further includes: when there is no intersection between the first music set and the second music set, inputting the learning task information into a preset music filtering model and outputting a reference music list; determining the historical preference score corresponding to each piece of music in the reference music list based on the user's historical music playback frequency, time decay factor, and scene relevance; and sorting the reference music list based on the historical preference score to construct the learning background music list.

[0014] In one implementation of this application, after adjusting the learning background music list based on the third music set, the method further includes: during the playback of the learning background music, comparing the obtained real-time learning state with preset state transition conditions, and regenerating the third music set when a change in user state is detected; determining the playback volume based on the changed user state, and automatically adjusting the player's volume; and playing the learning background music based on the regenerated third music set and the adjusted volume.

[0015] This application provides a music adaptive adjustment device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: perform task matching and task analysis based on learning scene information obtained by smart glasses to obtain learning task information; wherein the learning task information includes at least one of the following: learning task type, learning task duration, and learning task difficulty level; analyze and detect real-time learning behavior information obtained by smart glasses to obtain real-time learning status; and match corresponding background music in a music database based on the learning task information and / or real-time learning status to perform adaptive adjustment of background music.

[0016] The above-mentioned technical solutions adopted in this application embodiment can achieve the following beneficial effects: Firstly, this application embodiment acquires learning scenario information and user behavior information through smart glasses, achieving multi-dimensional data fusion and making music adjustment more intelligent and user-friendly. Secondly, this application embodiment senses the user's state in real time and adjusts the music accordingly, effectively alleviating distraction and fatigue, helping users maintain a positive learning attitude, and improving the overall learning experience. Furthermore, this application embodiment dynamically adjusts the music according to the type of learning task, task duration and difficulty, and the user's real-time state, providing personalized and precise music adjustment, reducing the interference caused by traditional fixed or preset music playback modes on the learning process, and improving learning efficiency. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0018] Figure 1 This is a schematic diagram of the system architecture that can be applied to the embodiments of this application;

[0019] Figure 2 A flowchart of a music adaptive adjustment method provided in this application embodiment;

[0020] Figure 3 This is a schematic diagram illustrating a method for determining the type of learning task provided in an embodiment of this application;

[0021] Figure 4 This is a schematic diagram illustrating a method for determining the duration of a learning task, as provided in an embodiment of this application.

[0022] Figure 5 This is a schematic diagram illustrating a method for determining the difficulty level of a learning task, provided in an embodiment of this application.

[0023] Figure 6 This is a schematic diagram illustrating a method for determining real-time learning status provided in an embodiment of this application;

[0024] Figure 7 This is a schematic diagram illustrating a method for adaptive adjustment of background music provided in an embodiment of this application;

[0025] Figure 8 A flowchart illustrating an adaptive music adjustment process provided in this application embodiment;

[0026] Figure 9 This is a schematic diagram of the structure of a music adaptive adjustment device provided in an embodiment of this application.

[0027] Figure label:

[0028] 101: Smart glasses, 102: Music database, 103: Music playback application, 104: Communication network, 105: Cloud. Detailed Implementation

[0029] This application provides a music adaptive adjustment method and apparatus.

[0030] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0031] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0032] Figure 1 This is a schematic diagram of the system architecture that can be applied to the embodiments of this application, such as... Figure 1 As shown, the music adaptive adjustment method can be applied to, for example... Figure 1 In the illustrated environment, the application environment may include smart glasses 101, a music database 102, a music playback application 103, a communication network 104, and a cloud 105. The smart glasses 101, worn by the user, are equipped with various learning environment information collection devices, such as cameras, to collect information about the learning environment within the user's field of vision. Additionally, the smart glasses 101 may be equipped with various user behavior collection devices, such as head posture sensors and biosensors, to collect real-time learning behavior information from the user. The smart glasses 101 uploads the collected learning environment information to the cloud 105 via the communication network 104. The cloud server analyzes the learning task type, duration, and difficulty level within the learning environment information. Based on the analysis results, it matches corresponding background music from the music database 102 and sends this background music to the music playback application 103 via the communication network 104, thereby controlling the music playback application 103 to play the background music. Subsequently, the smart glasses 101 transmits the user's learning behavior information acquired by the user behavior collection device to the cloud 105 via the communication network 104. The cloud server analyzes the user's learning behavior information to determine the user's learning status. Based on this status, it matches appropriate background music from the music database 102 and transmits it back to the music playback application 103 via the communication network 104 to control the application to update the playback tracks. Furthermore, the communication network 104 can also send playback volume control information from the cloud 105 to the smart glasses' companion application, allowing the application to adjust the volume of the playback device on the smart glasses 101.

[0033] Figure 2 A flowchart of a music adaptive adjustment method provided in this application embodiment is shown below. Figure 2 As shown, the music adaptive adjustment method includes the following steps S201-S203:

[0034] S201. Based on the learning scenario information obtained from the smart glasses, perform task matching and task analysis to obtain learning task information.

[0035] In one implementation of this application, the learning task information includes at least one of the following: learning task type, learning task duration, and learning task difficulty level. When determining the learning task type, the smart glasses have a built-in high-resolution camera that uses computer vision algorithms to capture real-time information about the learning environment in the user's field of vision. Image recognition and natural language processing technologies are then used to identify the type of learning activity the user is currently engaged in. The acquired learning environment information can include large blocks of text on a book, e-reader, or screen; textbook content; math problems; formulas; calculation processes on scratch paper; as well as the user's head posture, lip movements, and pronunciation actions during oral practice. Figure 3 This is a schematic diagram illustrating a method for determining the type of learning task provided in an embodiment of this application, such as... Figure 3 As shown, the method for determining the learning task type based on the acquired learning scenario information includes steps S301-S304:

[0036] S301. Perform image recognition processing on the image data in the learning scene information to extract visual feature information.

[0037] The acquired image data from the learning scenario is decomposed into multiple key regions, such as books, pens, the user's hands, head, and body parts, and these regions are used as nodes in a graph structure. The association weights between these nodes are calculated; for example, a graph attention model can be used to calculate the association weights between nodes. Emphasis is placed on core objects related to the learning scenario, such as identifying learning tools, learning materials, and the user's learning posture. Simultaneously, text regions within the images are located and labeled, and finally, various visual feature information is extracted from the images.

[0038] S302. Use natural language processing technology to perform text recognition on the text data in the learning scenario information and extract text feature information.

[0039] The acquired text data from the learning scenario is analyzed. If the text data comes from text regions in an image, it needs to be converted into editable text using optical character recognition (OCR). If it is existing electronic text, it is processed directly. Afterward, the text undergoes preprocessing such as word segmentation and stop word removal. Then, a word vector model is used to convert the text into vector representations, further extracting semantic features such as keywords, sentence structure, and topic orientation to form text feature information.

[0040] S303. Integrate visual feature information with text feature information to obtain key learning information.

[0041] The extracted visual and textual features are fused to obtain more comprehensive key learning information. During the fusion process, associations need to be established between the two types of features. For example, the visual feature of a math workbook is associated with keywords such as "function" and "geometry" in the text; the user's posture of looking down while writing is associated with the handwritten notes in the text. Feature fusion is performed through an attention-based fusion mechanism to obtain key learning information.

[0042] In one implementation of this application, when performing feature fusion through the attention fusion mechanism, firstly, for each set of associated visual and text features, the attention weight distribution is determined based on its relevance to the learning task. Based on the attention weight distribution, the visual features and text features are weighted and fused. The weighted visual features and text features are added or concatenated to form a comprehensive vector. By setting a threshold, the feature dimensions with weights higher than the threshold in the comprehensive vector are selected, thereby obtaining the key learning information.

[0043] S304. Match the key learning information with the pre-set learning task database to determine the learning task type.

[0044] This application embodiment pre-constructs a pre-built learning task database, which stores feature templates corresponding to different learning task types. The learning task types include at least one of the following: deep reading, language learning, mathematical problem-solving, and writing. For example, the feature template for deep reading may include: long text textbooks, user's focused reading posture (such as prolonged head-down posture, eyes fixed on the text area), etc.; the feature template for language learning may include: language learning application interfaces, word lists, dialogue texts, lip shapes and pronunciation movements when users practice speaking, etc.; the feature template for mathematical problem-solving may include: math textbooks, calculators, formula symbols in the text, user's posture during the writing calculation process, etc.; the feature template for writing may include: user's text input behavior on a computer or paper, brainstorming actions, and the writing interface, etc.

[0045] The fused key learning information is compared with the feature templates of each task type in the database, the similarity is calculated, and the task type corresponding to the feature template with the highest similarity is selected as the final determined current learning task type for the user.

[0046] Furthermore, if the learning task is in-depth reading, the system will prioritize recommending instrumental music and light music. These types of music typically have low rhythm and melodic complexity, which helps create a calm and focused atmosphere, preventing distractions. If the learning task is language learning, the system will select music with a moderate rhythm and clear, but not noisy, such as upbeat pieces, background music with a few vocal hums, or white noise of specific frequencies, to help users stay alert and aid in memory and pronunciation practice. If the learning task is mathematical problem-solving, the system tends to recommend classical music, such as forest birdsong or the sound of flowing water. This type of music helps build a rational and logical thinking space, promoting deep thinking and logical coherence when solving complex problems. If the learning task is writing, the system will recommend different types of music based on the nature of the writing content (e.g., creative writing, technical reports) and the user's current writing stage (e.g., brainstorming, drafting, revising). For example, more inspiring music might be needed during the brainstorming stage, while calmer music is needed when organizing thoughts.

[0047] Secondly, in this embodiment of the application, the duration of the learning task is determined based on the learning scenario information obtained by the smart glasses. Figure 4 This is a schematic diagram illustrating a method for determining the duration of a learning task, as provided in an embodiment of this application. Figure 4 As shown, the process of determining the duration of a learning task includes the following steps S401-S404:

[0048] S401. Determine the duration of the first reference learning task based on the amount of text content in the learning materials.

[0049] In one implementation of this application, the camera of the smart glasses can capture the content of the learning materials and analyze the number of questions, article length, number of chapters, etc. through image recognition and optical character recognition technologies, thereby making a preliminary estimate of the time required to complete the task.

[0050] Specifically, the total amount of text in the photographed learning materials can be quantified using indicators such as word count, paragraph count, or page count. Then, combined with preset text reading and comprehension rate indicators, such as the average number of words read per minute and the average processing time per paragraph, the time required to complete the first reference learning task for that text content can be calculated.

[0051] S402. Obtain the user-preset duration of the second reference learning task.

[0052] In one implementation of this application, the duration of a second reference learning task set by the user can be collected through a companion application of smart glasses. This duration is manually entered by the user according to their own plan, learning goals, or time schedule.

[0053] S403. In the historical database, identify historical learning data whose similarity values ​​with the key learning information meet the similarity criteria.

[0054] In one implementation of this application, based on the core features of the extracted current learning key information, such as the learning task type, the type of teaching materials involved, and user posture characteristics, a similarity algorithm is used to calculate its similarity value with historical data in the database. A similarity threshold is set to filter out historical learning data that meets the similarity criteria. The filtered historical learning data must include information such as the corresponding learning task duration and user completion status.

[0055] S404. Input the key learning information, the duration of the first reference learning task, the duration of the second reference learning task, and historical learning data into the preset duration prediction model to output the duration of the learning task corresponding to the current learning task.

[0056] The system integrates key learning information, the duration of the first reference learning task, the duration of the second reference learning task, and matched historical learning data to form an input feature set. This input feature set is then fed into a pre-set duration prediction model, which predicts the duration of the current learning task based on the correlations between the learning features.

[0057] During the training of the preset duration prediction model, the input samples are historical learning key information samples, historical first reference duration samples, historical second reference duration samples, and historical data statistical feature samples. The output is the actual completion duration sample of the corresponding historical learning task. The neural network model is trained to obtain the preset duration prediction model.

[0058] Secondly, in this embodiment of the application, the difficulty level of the learning task is determined based on the learning scenario information obtained by the smart glasses. Figure 5 This is a schematic diagram illustrating a method for determining the difficulty level of a learning task, as provided in an embodiment of this application. Figure 5 As shown, the process of determining the difficulty level of a learning task includes the following steps S501-S503:

[0059] S501. Determine the complexity level of the learning task based on the type of text content in the learning materials.

[0060] The complexity level of the learning task can be determined by identifying factors such as the density of technical terms, the complexity of formulas, and the amount of information in charts and graphs within the learning materials.

[0061] S502. In the historical database, identify the historical learning data corresponding to the complexity level of the learning task.

[0062] In the historical database, using the determined learning task complexity level as the search criterion, historical learning data of users completing tasks of the same complexity level is filtered out. This historical learning data must include the text type corresponding to the task, the user's operation process, and completion results data, such as task completion rate, error rate, and thinking time.

[0063] S503. Based on the completion results data corresponding to historical learning data, construct the user's learning profile and learning curve at the level of learning task complexity, so as to determine the difficulty level of the learning task according to the user profile and learning curve.

[0064] In one implementation of this application, a learning profile is constructed based on selected historical learning data. This learning profile includes: the user's strengths in learning, such as a high completion rate for a certain type of text; the user's weaknesses in learning, such as a high error rate for a certain type of question; and average thinking speed.

[0065] At the same time, with time as the horizontal axis and the results data, such as accuracy and efficiency, as the vertical axis, a learning curve is plotted to present the trend or fluctuation pattern of the user's learning task at this complexity level.

[0066] Furthermore, by comprehensively analyzing user profiles and learning curves, the actual difficulty of the current task for the user is determined. Specifically, if the user profile shows a high completion rate and low error rate in tasks of similar complexity, and the curve shows a continuous upward trend, it indicates that the current task is relatively easy; if the profile reflects a high error rate and long thinking time in similar tasks, and the curve fluctuates greatly or rises slowly, then the current task is relatively difficult. Combining this with preset difficulty grading standards, such as easy, medium, and hard, the difficulty level of the learning task is finally determined.

[0067] This application employs different music adjustment strategies for varying learning task durations and difficulty levels. When the system anticipates a long and challenging learning task, it generates a series of soothing and layered music tracks. This music adjustment is gradual, starting with calm, background music and transitioning to music with moderate rhythmic variations that won't disrupt concentration. For example, it might begin with pure ambient sounds, gradually adding gentle melodies, followed by instrumental music with subtle beats, ensuring the music always serves as a supplement rather than a distraction. This gradual adjustment helps users maintain focus and alleviate cognitive fatigue during extended learning sessions. For short, low-difficulty tasks, the system may select more upbeat and energizing music to quickly stimulate the user's learning interest and efficiency.

[0068] S202. Analyze and detect the real-time learning behavior information obtained by the smart glasses to obtain the real-time learning status.

[0069] In one implementation of this application, user learning behavior information obtained through smart glasses is used to detect in real time whether the user is distracted or fatigued, and the generated background music is automatically adjusted according to the user's current state. For example, when the user is distracted: classical or light music is generated to help the user concentrate and alleviate distraction. When the user is fatigued: upbeat but not overly complex pop music is generated to help the listener shift their attention away from the negative emotions of fatigue and face the fatigue state with a more positive attitude. At the same time, the upbeat rhythm can also invigorate the spirit and alleviate physical fatigue to a certain extent. Figure 6 This is a schematic diagram illustrating a method for determining real-time learning status provided in an embodiment of this application, such as... Figure 6 As shown, the method for determining the real-time learning state includes steps S601-S603:

[0070] S601. Obtain real-time learning behavior information sent by different information collection devices on the smart glasses.

[0071] Real-time learning behavior information includes at least one of the following: eye tracking information, head posture information, facial expression information, and physiological characteristic parameters.

[0072] In one implementation of this application, the information acquisition device equipped with smart glasses acquires real-time learning behavior information of the user during the learning process. For example, eye-tracking sensors capture eye-tracking information, including gaze position, fixation duration, and blink frequency; gyroscopes and accelerometers record head posture information, such as head rotation angle, tilt amplitude, and stability duration, and monitor fatigue postures such as head drooping and frequent nodding; a camera captures and extracts facial expression information, including features such as the curvature of the corners of the mouth, frowning frequency, and eye expression changes, and identifies fatigue signs such as drooping eyelids, frequent eye rubbing, and eye congestion; and biosensors collect physiological characteristic parameters, such as heart rate changes and skin conductance. After collection, the data is categorized and stored to ensure that each type of data is independent and complete.

[0073] S602. Compare different types of real-time learning behavior information with their respective state thresholds, and determine the state coefficient values ​​corresponding to each type of real-time learning behavior information based on the comparison difference.

[0074] In one implementation of this application, for each type of real-time learning behavior information after classification, a preset state threshold is compared to calculate the corresponding state coefficient value. For example, for eye-tracking information, if the duration of gaze outside the learning material exceeds the distraction state threshold, or the gaze duration is lower than the focus state threshold, a distraction coefficient between 0 and 1 is assigned based on the degree of exceeding or falling below the threshold. For physiological characteristic parameters, if the heart rate fluctuation amplitude or blink frequency exceeds the fatigue state threshold, a fatigue coefficient between 0 and 1 is assigned based on the degree of deviation. Each type of information is compared using similar logic to obtain a state coefficient value reflecting the degree of distraction or fatigue in that dimension; the closer the coefficient is to 1, the more pronounced the corresponding state.

[0075] S603. Based on the preset weight allocation parameters, real-time learning behavior information, and state coefficient values, the real-time learning state is obtained.

[0076] In one implementation of this application, real-time learning behavior information of the type required for distraction detection is filtered from the real-time learning behavior information to construct a distraction detection information set. Based on the weight allocation parameters and state coefficient values ​​corresponding to the distraction detection information set, the real-time learning behavior information in the distraction detection information set is weighted to obtain the user's distraction state.

[0077] Specifically, from real-time learning behavior information, information types highly correlated with distraction states are selected to construct a distraction detection information set. This distraction detection information set includes at least one of the following: eye-tracking information, head posture information, and facial expression information. For each type of information in the distraction detection information set, its corresponding weight allocation parameters are called, and weighted processing is performed in conjunction with the calculated state coefficient values. For example, the distraction coefficient of eye-tracking is multiplied by its weight, the distraction coefficient of head posture is multiplied by its corresponding weight, and then all weighted results are summed to obtain a comprehensive score for the distraction state. If this score exceeds a preset distraction judgment threshold, the user is determined to be in a distracted state; the higher the score, the more severe the distraction.

[0078] When the system detects that the user is distracted, it immediately adjusts the music style, such as switching to classical music, instrumental music, or light music. These types of music are typically characterized by a tight structure, smooth melody, and steady rhythm, which can help the user refocus their attention and guide their thinking back to the learning task.

[0079] In one implementation of this application, real-time learning behavior information of the type required for fatigue detection is selected from the real-time learning behavior information to construct a fatigue detection information set. Based on the weight allocation parameters and state coefficient values ​​corresponding to the fatigue detection information set, the real-time learning behavior information in the fatigue detection information set is weighted to obtain the user's fatigue state.

[0080] Specifically, from real-time learning behavior information, information types highly correlated with fatigue state are selected to construct a fatigue detection information set. This fatigue detection information set includes at least one of the following: eye features, head posture, and physiological characteristic parameters. Among them, eye features can be signs of fatigue such as drooping eyelids, frequent eye rubbing, and eye congestion; head posture can be fatigue postures such as drooping head and frequent nodding.

[0081] Furthermore, for each type of information in the fatigue detection data set, its corresponding weighting parameters are invoked and weighted in conjunction with existing state coefficient values. For example, the fatigue coefficient of physiological characteristics is multiplied by a weight, the fatigue coefficient of blinking frequency is multiplied by a corresponding weight, and all weighted results are summed to obtain a comprehensive score for the fatigue state. If this score exceeds a preset fatigue judgment threshold, the user is determined to be in a fatigue state; the higher the score, the more pronounced the fatigue.

[0082] When user fatigue is detected, the system generates upbeat but not overly complex pop, light rock, or electronic music. This type of music typically features a brisk rhythm and upbeat melody, helping listeners shift their attention away from the negative emotions of fatigue and fostering a positive mindset. At the same time, the upbeat rhythm can invigorate the spirit and alleviate physical fatigue to some extent, while avoiding overly intense music that could cause new fatigue or distraction.

[0083] S203. Based on learning task information and / or real-time learning status, match the corresponding background music in the music database and perform adaptive adjustment of the background music.

[0084] This application embodiment dynamically adjusts the music according to the learning scenario, task duration and difficulty, and the user's real-time status, providing personalized and precise music adjustment services. Figure 7 This is a schematic diagram of a method for adaptive adjustment of background music provided in an embodiment of this application, as shown below. Figure 7 As shown, the method for adaptive adjustment of background music includes the following steps S701-S705:

[0085] S701. Based on the learning task type, determine the first music set in the music database.

[0086] Based on the identified learning task type, the first music set suitable for that task type is selected from the music database. For example, if the task is identified as "deep reading" and the user is currently in a good state, then instrumental music or light music will be matched.

[0087] S702. Based on the learning task duration and difficulty level, construct a music characteristic trend curve, and determine the second music set in the music database based on the music characteristic trend curve; wherein, the music characteristic trend curve is divided into multiple playback stages based on the changes in music characteristics.

[0088] In one implementation of this application, a music characteristic trend curve is constructed based on the learning task duration and difficulty level to pre-adjust the background music. For example, the learning process can be divided into multiple playback stages based on duration, such as the first 20% as the introductory stage, the middle 60% as the focus stage, and the last 20% as the concluding stage. Then, a music characteristic change pattern is set for each stage in conjunction with the difficulty level; for example, the music rhythm can be slightly faster in the focus stage of low-difficulty tasks, while it needs to be smoother for high-difficulty tasks; when the duration is longer, the transition of music characteristics between stages needs to be more natural. Based on this curve, music that meets the characteristic requirements of each stage is selected from a music database and compiled into a second music set to ensure that the music adapts to the task requirements as the learning progresses.

[0089] For example, for long and challenging tasks, the music should start calm and low-energy, gradually introducing richer melodies and moderate rhythms, but always maintaining the principle of not interfering with thinking.

[0090] S703. During multiple playback stages, a list of background music for learning is determined based on the intersection of the first music set and the second music set, and then played.

[0091] In one implementation of this application, the intersection of the first music set and the second music set is determined in each of the multiple playback stages divided by the music characteristic trend curve, resulting in candidate music for each stage. These candidate music tracks are then integrated in stage order to form a learning background music list, ensuring that the music in the list not only matches the learning task type but also meets the music characteristic requirements of each stage. Subsequently, the background music is automatically played according to the list order, creating a suitable learning atmosphere for the user.

[0092] In one implementation of this application, when the first music set and the second music set do not overlap, the learning task information is input into a pre-set music filtering model, which outputs a reference music list. Based on the user's historical music playback frequency, time decay factor, and scene relevance, the historical preference score corresponding to each piece of music in the reference music list is determined. The reference music list is then sorted based on the historical preference scores to construct a learning background music list.

[0093] When the first and second music sets do not overlap, the learning task information is input into a pre-defined music selection model. The input samples for training this model can be: learning task type samples, duration samples, and difficulty level samples; the output samples can be a list of music that meets the criteria. This pre-defined music selection model analyzes the correlation between task information and music characteristics, such as style, rhythm, and melodic complexity, to select music suitable for the current learning scenario from a music database, generating a reference music list. Based on the user's historical music playback data, a historical preference score is calculated for each song in the reference music list. Specifically, the historical playback frequency of each song is calculated; the higher the frequency, the higher the base score. A time decay factor is introduced, giving higher weight to recently played music and appropriately reducing the weight of older music to avoid outdated preferences affecting the results. Simultaneously, scenario relevance is considered, i.e., the number of times the music has been played in similar learning scenarios in the past; the higher the relevance, the greater the score bonus. The values ​​of these three dimensions are integrated according to a preset ratio to obtain the historical preference score for each song, reflecting the user's potential acceptance of that music. Based on the calculated historical preference scores, the music in the reference music list is sorted in descending order, with the highest-scoring music at the top and so on down. After sorting, the sorted music is allocated to each stage based on the learning task duration and the divided playback phases, ensuring that the number of music tracks in each stage matches the duration and that the overall style remains consistent.

[0094] S704. Based on the real-time learning state, determine the third music set in the music dataset.

[0095] In one implementation of this application, a third music set that can improve the real-time learning state is selected from the music database based on the real-time learning state.

[0096] For example, when a user is distracted, the system will immediately adjust the music style, such as switching to classical music, instrumental music, or light music with a rigorous structure, smooth melody, and steady rhythm. The volume can be appropriately lowered to reduce external interference, help the user refocus their attention, and guide their thinking back to the learning task.

[0097] When a user is fatigued, the system generates upbeat but not overly complex pop music, light rock, electronic music, or instrumental pieces with positive elements. This type of music typically features a brisk rhythm and upbeat melody, helping listeners shift their attention away from the negative emotions of fatigue and fostering a positive mindset. At the same time, the upbeat rhythm can invigorate the spirit and alleviate physical fatigue to some extent, while avoiding overly intense or complex music that could introduce new cognitive load.

[0098] If the user is anxious or irritable, the system can switch to soothing natural sounds, meditation music, or low-frequency white noise to help the user calm down.

[0099] For users who maintain high concentration for extended periods, the system can maintain the current music style or make minor adjustments to avoid auditory fatigue.

[0100] S705. Adjust the learning background music list based on the third music set.

[0101] If the system detects that the user is distracted or fatigued, it will select suitable music from the third music collection to replace the music in the corresponding stage of the list, or adjust the music playback parameters, such as volume and tempo.

[0102] In one implementation of this application, during the playback of background music for learning, the acquired real-time learning state is compared with preset state transition conditions. If a change in user state is detected, a third music set is regenerated. Based on the changed user state, the playback volume is determined, and the player's volume is automatically adjusted. The background music for learning is then played based on the regenerated third music set and the adjusted volume.

[0103] Specifically, during the background music playback, the system continuously acquires the user's real-time learning status and compares it against preset status transition conditions. These conditions include trigger thresholds for transitioning from focus to distraction and criteria for determining transitioning from alertness to fatigue. By comparing the conditions, the system determines whether the user's current state has transitioned. If the transition conditions are met, the subsequent music adjustment process is initiated; otherwise, the current playback status is maintained.

[0104] When a change in user state is detected, a third music set is regenerated based on the new state. For example, when a user transitions from distraction to focus, soothing music that reinforces focus is selected; when a user transitions from fatigue to alertness, upbeat but not overly stimulating music is retained.

[0105] Furthermore, based on the user's transformed state, the corresponding playback volume is determined in conjunction with preset volume adjustment rules. For example, when a user is distracted and needs to increase their attention, the volume can be set slightly higher than normal to moderately stimulate the hearing; when a user is fatigued and needs to relieve fatigue, the volume is adjusted slightly lower than normal to avoid excessive volume that could worsen fatigue.

[0106] The background music player automatically adjusts to the set playback volume, ensuring a smooth transition to the target value and preventing sudden volume changes from disturbing the user. Simultaneously, it selects music suitable for the current stage from a newly generated third music set, replacing the corresponding tracks in the original playlist to ensure synchronized music switching and volume adjustment.

[0107] In this embodiment, the adjusted music is played to the user through the smart glasses' built-in miniature speaker or via Bluetooth-connected headphones. The system continuously monitors the user's status, forming a closed-loop feedback loop to continuously optimize the music adjustment effect. All adjustment processes strive for smoothness and seamlessness, avoiding abrupt music switching that could disturb the user. The system can also record the effect of each music adjustment and incorporate it into the training data of a machine learning model, further improving the accuracy and intelligence of adaptive adjustment.

[0108] Figure 8 A flowchart illustrating an adaptive music adjustment process provided in this application embodiment is shown below. Figure 8 As shown, this embodiment of the application collects learning scene information using the visual recognition technology of smart glasses. By identifying the features of the learning scene, it matches the corresponding learning task type. Then, by collecting and analyzing the learning content, the user's learning posture, and physiological characteristic parameters, it determines the duration and difficulty level of the current learning task. Next, by acquiring the user's behavioral information during learning, it detects the user's real-time state and adjusts the background music based on that state. For example, if the user is distracted, the system automatically switches to classical or light music; if the user is fatigued, the system automatically switches to upbeat pop music; if the user is in a normal learning state, it continues to use the background music matched by the learning task type, duration, and difficulty level. Finally, through music playback and feedback, the music adjustment effect is continuously optimized.

[0109] Figure 9 This is a schematic diagram of the structure of a music adaptive adjustment device provided in an embodiment of this application. Figure 9 As shown, a music adaptive adjustment device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform any of the above-described music adaptive adjustment methods.

[0110] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0111] The above descriptions are merely embodiments of this application and are not intended to limit the scope of this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions in the embodiments of this application.

Claims

1. A music adaptive adjustment method, characterized in that, The method includes: Based on the learning scenario information obtained by the smart glasses, task matching and task analysis are performed to obtain learning task information; wherein, the learning task information includes at least one of the following: learning task type, learning task duration, and learning task difficulty level; The real-time learning behavior information acquired by the smart glasses is analyzed and detected to obtain the real-time learning status; Based on the learning task information and / or the real-time learning status, match the corresponding background music in the music database and perform adaptive adjustment of the background music. The step of performing task matching and analysis based on the learning scenario information obtained from the smart glasses to obtain learning task information specifically includes: Image recognition processing is performed on the image data in the learning scene information to extract visual feature information; Natural language processing technology is used to perform text recognition on the text data in the learning scenario information and extract text feature information. The visual feature information and the text feature information are fused to obtain key learning information; The key learning information is matched against a pre-set learning task database to determine the type of learning task. The duration of the first reference learning task is determined based on the amount of text content in the learning materials; Obtain the user-preset duration of the second reference learning task; In the historical database, identify historical learning data whose similarity values ​​meet the similarity criteria with the aforementioned key learning information; The learning key information, the duration of the first reference learning task, the duration of the second reference learning task, and the historical learning data are input into a preset duration prediction model to output the duration of the learning task corresponding to the current learning task. The complexity level of the learning task is determined based on the type of text content in the learning materials; In the historical database, the historical learning data corresponding to the complexity level of the learning task is determined; Based on the completion results data corresponding to the historical learning data, a learning profile and learning curve corresponding to the user are constructed under the complexity level of the learning task, so as to determine the difficulty level of the learning task according to the learning profile and the learning curve. The step of matching corresponding background music in the music database based on the learning task information and / or the real-time learning status, and performing adaptive adjustment of the background music, specifically includes: Based on the learning task type, a first music set is determined in the music database; Based on the learning task duration and difficulty level, a music characteristic trend curve is constructed, and a second music set is determined in the music database based on the music characteristic trend curve; wherein, the music characteristic trend curve is divided into multiple playback stages based on changes in music characteristics; In the multiple playback stages, a list of background music for learning is determined based on the intersection of the first music set and the second music set, and then played. Based on the real-time learning state, a third music set is determined in the music dataset; The learning background music list is adjusted based on the third music set.

2. The method according to claim 1, characterized in that, The visual feature information includes at least one of the following: learning tools, learning materials, user learning posture, and text data generated during user learning; The learning task types include at least one of the following: in-depth reading, language learning, mathematical problem-solving, and writing.

3. The method according to claim 1, characterized in that, The completion results data include at least the task completion rate, task error rate, and thinking time.

4. The method according to claim 1, characterized in that, The step of analyzing and detecting the real-time learning behavior information acquired by the smart glasses to obtain the real-time learning state specifically includes: The system acquires real-time learning behavior information sent by different information acquisition devices on the smart glasses; wherein the real-time learning behavior information includes at least one of the following: eye tracking information, head posture information, facial expression information, and physiological feature parameters. The different types of real-time learning behavior information are compared with the corresponding state thresholds, and the state coefficient values ​​corresponding to each type of real-time learning behavior information are determined based on the comparison difference. The real-time learning state is obtained based on the preset weight allocation parameters, the real-time learning behavior information, and the state coefficient value. The real-time learning state includes a distracted state and a fatigued state.

5. The method according to claim 4, characterized in that, The process of obtaining the real-time learning state based on preset weight allocation parameters, the real-time learning behavior information, and the state coefficient value specifically includes: From the real-time learning behavior information, the types of real-time learning behavior information required for distraction detection are selected to construct a distraction detection information set; Based on the weight allocation parameters corresponding to the distraction detection information set and the state coefficient value, the real-time learning behavior information in the distraction detection information set is weighted to obtain the user's distraction state. From the real-time learning behavior information, the types of real-time learning behavior information required for fatigue detection are selected to construct a fatigue detection information set; Based on the weight allocation parameters corresponding to the fatigue detection information set and the state coefficient value, the real-time learning behavior information in the fatigue detection information set is weighted to obtain the user's fatigue state.

6. The method according to claim 1, characterized in that, Before determining and playing the background music list for learning, the method further includes: If there is no intersection between the first music set and the second music set, the learning task information is input into the preset music filtering model, and a reference music list is output. Based on the user's historical music playback frequency, time decay factor, and scene relevance, the historical preference score corresponding to each piece of music in the reference music list is determined. The reference music list is sorted based on the historical preference scores. After sorting, the sorted music is allocated to each stage based on the learning task duration and the divided playback stages, ensuring that the number of music tracks in each stage matches the duration, thus constructing the learning background music list.

7. The method according to claim 1, characterized in that, After adjusting the learning background music list based on the third music set, the method further includes: During the background music playback, the obtained real-time learning state is compared with the preset state transition conditions. If a change in the user's state is detected, the third music set is regenerated. Based on the converted user state, the playback volume is determined, and the player's volume is automatically adjusted. Based on the regenerated third music set and the adjusted volume, background music is played for learning.

8. A music adaptive adjustment device, characterized in that, The device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to: perform the music adaptive adjustment method according to any one of claims 1-7.

Images