An individualized music intervention system and method

Abstract

The application relates to the technical field of psychological intervention, and discloses an individualized music intervention system and method. In the system, a music preference acquisition module is used to acquire music preference information of a subject; a matching rule setting module is used to determine a matching rule of music preference and intervention demand according to an intervention target; a music material acquisition module is used to acquire a plurality of required music materials according to the music preference information and the matching rule; a sequence arrangement module is used to combine and arrange the music materials to form an intervention audio sequence; and a playing control module is used to periodically provide the subject with the intervention audio sequence in a preset intervention period. The application can provide precise and individualized music intervention and improve intervention effect.

Description

Technical Field

[0001] This invention relates to the field of psychological intervention technology, and in particular to an individualized music intervention system and method. Background Technology

[0002] Existing music intervention techniques have many shortcomings in practical application. Traditional music interventions rely heavily on the personal experience and subjective judgment of the interventionist to select music materials. This approach lacks standardized selection criteria and scientific emotion regulation strategies, resulting in poor repeatability of intervention effects and significant differences in intervention levels among different interventionists. Furthermore, interventionists struggle to accurately grasp the individualized needs of participants, neglecting the significant impact of individual differences on intervention outcomes.

[0003] In clinical practice, there is a general lack of effective methods for systematically applying subjects' music preference information to the development of intervention programs.

[0004] Existing music intervention programs are mostly static and fixed, using the same music throughout the intervention period, and cannot be dynamically adjusted according to the changes in the subject's condition at different stages of the intervention. This inflexible intervention approach is difficult to adapt to the real-time needs of the subjects, limiting further improvement in the intervention effect.

[0005] With the development of artificial intelligence technology, music generation technology has made significant progress, but its application in the field of clinical music intervention is still in the exploratory stage.

[0006] In addition, the efficiency of existing music intervention programs needs to be improved.

[0007] In summary, existing music intervention techniques have significant shortcomings in areas such as the systematic collection and application of preference information, the development of individualized intervention plans, the scientific implementation of emotion regulation mechanisms, the clinical application of artificial intelligence technology, and the automation of the intervention process. Summary of the Invention

[0008] The purpose of this invention is to propose an individualized music intervention system and method to solve the technical problem that existing music interventions lack systematic emotion regulation strategies.

[0009] To achieve the above objectives, the present invention provides a personalized music intervention system, comprising: The music preference acquisition module is used to obtain the music preference information of the subjects; The matching rule setting module is used to determine the matching rules between music preferences and intervention needs based on the intervention objectives; The music material acquisition module is used to acquire multiple music materials that meet the requirements based on the music preference information and the matching rules. The sequence arrangement module is used to combine and arrange the music materials to form an intervention audio sequence; The playback control module is used to periodically provide the intervention audio sequence to the subject within a preset intervention period.

[0010] Optionally, the matching rule setting module includes an emotion curve design unit, used to design an emotion arousal curve based on the emotion entrainment theory. The emotion arousal curve specifies the trajectory of changes in musical emotion during the intervention period and requires that the emotional expressiveness of the musical material matches the corresponding time point of the emotion arousal curve.

[0011] Optionally, the emotion curve design unit is configured as follows: For anxiety or depression disorders, an emotional arousal curve is designed that starts from a negative or low positive emotional level, gradually rises to a peak of positive emotion, and then falls back to a moderate level of positive emotion. For sleep disorders, an emotional arousal curve was designed, which initially raises positive emotions to a moderate level and then gradually decreases them to a calm state in the later stages.

[0012] Optionally, the music material acquisition module further includes an audio processing unit, which, for sleep disorders, is configured to perform at least three technical processes on the music material: Gradually reduce the playback speed of the music; Simplify the complexity of rhythm and melody; Increase the volume transition time between music tracks; Amplitude limiting measures were implemented. Prioritize music materials with regular repetition characteristics.

[0013] Optionally, the music preference acquisition module is configured to test the target group using a music style preference scale and select music style types with higher scores as preference references; The music material acquisition module includes a music material database and a professional music selection interface. It selects music materials that meet the matching rules from the music material database, taking into account the aesthetic value of the music and its ability to evoke positive imagery.

[0014] Optionally, the matching rule setting module is configured to directly map the features of each dimension in the music preference information to the corresponding feature requirements of the music material.

[0015] Optionally, the music preference acquisition module is configured to repeatedly acquire the subject's current music preference information at multiple time points within the preset intervention period. The music preference information includes at least three dimensions: music mood, music style, music intensity, instrument type, music tempo, and degree of music variation.

[0016] Optionally, the music material acquisition module includes: A preference conversion unit is used to convert the music preference information into music description statements; A music matching unit is used to generate or retrieve multiple candidate music materials based on the music description statement; The scoring collection unit is used to collect the subjects' preference ratings for the candidate music materials; The intelligent filtering unit is used to filter music materials with higher scores based on the preference rating.

[0017] Optionally, the music matching unit includes an artificial intelligence music generation subunit and / or a music database retrieval subunit; The AI ​​music generation subunit is configured as follows: Call the AI-powered music generation platform via API interface; Generate multiple candidate music materials; and Store the generated music file; The music database retrieval subunit is configured as follows: Perform a multi-condition search in the music material database based on the aforementioned music description statement; Calculate the match score of the retrieved music; and Sort by matching score and select multiple candidate music materials.

[0018] This invention also provides a personalized music intervention method, comprising: Obtain the subjects' music preference information; Determine the matching rules between music preferences and intervention needs based on the intervention goals; Based on the music preference information and the matching rules, multiple music materials that meet the requirements are obtained; and The music materials are combined and arranged to form an interventional audio sequence; The matching rules include: Based on the theory of emotional entrainment, a preset emotional arousal curve is designed. The emotional arousal curve defines the trajectory of changes in musical emotions during the intervention period, requiring that the emotional expressiveness of the musical material matches the corresponding time point of the emotional arousal curve. or The features of each dimension in the music preference information are directly mapped to the corresponding feature requirements of the music material.

[0019] Compared with existing technologies, this invention acquires the subject's music preference information and establishes matching rules based on the intervention goals, systematically matching music preference information with intervention needs. The selected or generated music materials conform to the subject's personal preferences. Furthermore, intervention audio sequences are provided periodically, and multiple music materials are combined and arranged to form a complete intervention audio sequence, avoiding the monotony of a single music material and enhancing the attractiveness and sustainability of the intervention.

[0020] This invention further provides two preferred methods for implementing matching rules. The first method designs an emotion arousal curve based on the emotion entrainment theory, guiding the subject's emotional state to gradually improve through a scientific emotion regulation path, particularly by designing differentiated emotion curves for different intervention goals. The second method directly maps multiple dimensions of music preference information to music material requirements, achieving more flexible and precise individualized customization. It responds in real-time to changes in the subject's current preferences, generating or retrieving music materials that best meet their needs, significantly improving personalization and the accuracy of the intervention. Attached Figure Description

[0021] Figure 1 This is a block diagram of a personalized music intervention system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of music-based emotional guidance for anxious and depressed subjects in an embodiment of the present invention; Figure 3 This is a schematic diagram of music-based emotional guidance for subjects with sleep disorders in an embodiment of the present invention; Figure 4 This is a flowchart of an individualized music intervention method according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating how to obtain multiple music materials that meet the requirements in an embodiment of the present invention. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. It should be noted that those skilled in the art can make appropriate modifications to the present invention without affecting its intended benefits. Therefore, the following description should be considered as a general understanding to those skilled in the art and is not intended to limit the invention.

[0023] The invention is described more specifically by way of example in the following paragraphs with reference to the accompanying drawings. The advantages and features of the invention will become clearer from the following description. It should be noted that the drawings are in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of the invention.

[0024] Example 1 This invention provides a personalized music intervention system; please refer to [the relevant documentation]. Figure 1 The system includes: a music preference acquisition module, a matching rule setting module, a music material acquisition module, a sequence arrangement module, and a playback control module.

[0025] In this embodiment, the music preference acquisition module is used to acquire the subject's music preference information; the matching rule setting module is used to determine the matching rules between music preference and intervention needs according to the intervention goal; the music material acquisition module is used to acquire multiple music materials that meet the requirements according to the music preference information and the matching rules; the sequence arrangement module is used to combine and arrange the music materials to form an intervention audio sequence; and the playback control module is used to periodically provide the intervention audio sequence to the subject within a preset intervention period.

[0026] This system adopts an architecture combining cloud servers and mobile terminals. The cloud server hosts backend business logic, databases, API interfaces, music libraries, etc. The mobile terminal provides an interactive interface for participants, allowing them to interact with the system through applications on smartphones or tablets. The application provides functions such as questionnaire completion, music listening and rating, audio playback, and scale assessment. The mobile terminal and cloud server are connected via the internet (Wi-Fi or mobile data network).

[0027] In a specific example, the music preference acquisition module obtains preference information from the subjects through the application interface and sends the data to the matching rule setting module on the cloud server.

[0028] After receiving the preference information, the matching rule setting module generates matching rules based on the intervention target and the preset matching strategy, and then passes the matching rules to the music material acquisition module.

[0029] The music acquisition module selects music from the music database or calls the AI ​​platform to generate music according to the matching rules, obtains music materials that meet the requirements, and passes the material data to the sequence arrangement module.

[0030] The sequence arrangement module combines and arranges music materials into a complete intervention audio sequence, and transmits the audio information to the playback control module.

[0031] The playback control module provides audio playback services to subjects through an application.

[0032] Depending on the different strategies adopted by the matching rule setting module, the system has two main implementation modes, which are described in detail below.

[0033] Example 2 In this embodiment, when the matching rule setting module adopts the emotional arousal curve strategy, the specific implementation methods of each module of the system of the present invention are as follows. This implementation mode is particularly suitable for improving anxiety, depression, and sleep disorders.

[0034] In this embodiment, the music preference acquisition module is specifically configured to test the target group through a music style preference scale in order to understand the basic music preferences of the subject group.

[0035] Specifically, the STOMPS-R (Short Test of Music Preferences-Revised) scale was used to conduct a group test on the participants. The STOMPS-R scale assesses the participants' preference for 28 different styles of music and divides them into 4 categories. Participants rated their preference for music styles from 1 to 7, ranging from very dislike to very like.

[0036] In one specific example, the system presents a list of 28 music styles, each accompanied by a brief description and a link to listen to sample music. The system receives the participants' ratings in real time and stores them in a database. Once a sufficient sample size is collected, the system automatically calculates the average rating for each style and identifies the music style types with higher ratings.

[0037] In one specific example, participants experiencing anxiety, depression, and insomnia underwent the STOMP-R test. The results showed that participants gave higher scores to two music categories: Reflective & Complex and Upbeat & Conventional. The average score for the Reflective & Complex style was 5, while the average score for the Upbeat & Conventional style was 4.7, indicating that both styles were similar or met with a high level of liking.

[0038] Reflective and Complex music encompasses eight genres: bluegrass, blues, classical, folk, world music, jazz, new age, and opera. This genre exhibits a wide range of positive and negative emotions, is slower in tempo than other genres, primarily uses acoustic instruments, and is considered complex in nature, with high levels of both positive and negative emotions but low energy levels.

[0039] Optimistic and traditional music encompasses six styles: country music, gospel music, religious music, oldies, pop music, and soundtracks / themes. This category primarily expresses positive emotions, features simple structures, and moderate energy. Its musical attributes are considered simple and direct, with moderate positive emotions and lower levels of negative emotions and energy.

[0040] To understand the impact of musical elements rather than lyrics on participants, the system's selection criteria were set to use only purely instrumental pieces. Therefore, the final selections primarily featured instrumental music in genres such as classical, new age, pop, soundtracks / theme songs, and jazz.

[0041] Furthermore, the music preference acquisition module is used to pass the statistical analysis results to the matching rule setting module as a preference reference for subsequent music material selection.

[0042] In this embodiment, the subjectivity of music selection was addressed by testing group preferences using a standardized scale, providing a scientific basis for the selection of subsequent music materials and increasing the acceptance of music intervention.

[0043] In this embodiment, the matching rule setting module includes an emotion curve design unit, which is used to design an emotion arousal curve based on the emotion entrainment theory. The emotion arousal curve specifies the trajectory of music emotion changes during the intervention period and requires that the emotional expressiveness of the music material matches the corresponding time point of the emotion arousal curve.

[0044] The specific components of the emotion curve design unit include: Intervention Target Identification Subunit: Automatically identifies intervention targets based on the subject's diagnostic information, such as anxiety disorders, depressive disorders, or sleep disorders.

[0045] In a specific example, the intervention target identification subunit can read the subject's basic information and diagnostic records from the system database to determine the type of emotion curve to be applied.

[0046] Emotion curve template library: Pre-stored emotion curve templates for different intervention goals.

[0047] In a specific example, the database stores multiple clinically validated emotion curve templates, each containing information such as a timeline, emotion value sequence, and curve description.

[0048] Curve parameter adjustment subunit: Allows for fine-tuning of the curve template based on the specific circumstances of the subject.

[0049] The curve parameter adjustment sub-unit provides a parameter adjustment interface, which can be used to modify parameters such as starting point emotion value, peak emotion value, fallback value, and duration of each stage to adapt to individual differences.

[0050] Timeline Mapping Subunit: Maps the emotion curve to a specific timeline. This subunit transforms the abstract emotion curve into a concrete time-emotion value correspondence, such as: [(0min, -1 points), (5min, 0 points), (10min, 3 points), (15min, 5 points), (20min, 7 points), (25min, 5 points), (30min, 3 points)].

[0051] The theoretical basis of the emotional entrainment theory is that musical emotion is an innate biological function of the organism. When the body is in a musical emotional state, the level of physiological arousal decreases, the state of tension is relieved, and the physiological state is relatively close to homeostasis. Homeostasis is the optimal level of physiological arousal, enabling the organism to maintain the normal activities of various organs and tissues with minimal physiological energy. The basic mechanism of musical emotion is mainly the tension-relaxation movement change at a low physiological level.

[0052] From the perspective of neurophysiological entrainment theory, music can induce an entrainment response in humans. Entrainment is a form of communication between organized, interconnected neurons. During the listening to musical rhythms, four levels of entrainment occur: spontaneous perceptual, physiological, motor, and social. From a physiological entrainment perspective, the autonomic nervous system, controlled by the sympathetic and parasympathetic nervous systems, exhibits an entrainment response to externally perceived rhythms, primarily involving respiration and heart rate. Respiratory rate unconsciously adjusts according to the tempo of the rhythm. For example, slow, soothing music can reduce heart rate, respiratory rate, and blood pressure.

[0053] Furthermore, the Mood Induction Technique is a professional method for improving mood. It involves playing audio that matches the participant's mood and conveys a range of emotional states, gradually changing the music while maintaining awareness of the participant's mood. The aim is to change the mood through the music, gradually guiding the participant into a better or more adaptive state.

[0054] In individualized music intervention clinical practice, the term "carrying" is also translated as "carrying". It advocates that not only should the music be consistent with the client's emotions at the beginning of the intervention, but also emphasizes that the mood of the music should be gradually changed during the intervention, using music to "carry" the client's emotions toward the predetermined intervention goal.

[0055] In this embodiment, the emotion curve design unit is configured to design differentiated emotion arousal curves for different types of obstacles.

[0056] Specifically, the design of emotional arousal curves for anxiety or depressive disorders: Please refer to Figure 2 The emotion curve design unit is configured to design an emotion arousal curve that starts from a negative or low positive emotion level, gradually rises to a positive emotion peak, and then falls back to a moderate positive emotion level.

[0057] Clinical subjects with depression and anxiety often experience low mood; therefore, the initial music should match the subject's current emotional state to establish resonance. The energy and emotion of the music should then be gradually increased, guiding the subject's mood towards a positive direction. Finally, the music should be appropriately lowered to allow the subject to adjust to their daily life after listening to the music, avoiding an excessive emotional drop.

[0058] In a specific example, the emotion quantification scoring system ranges from -2 to +10 points, where negative values ​​represent negative emotions and positive values ​​represent positive emotions. The emotion arousal curve changes as follows: starting from -1 point (calm with slight sadness), it gradually increases to +7 points (joyful, grand, and affectionate), and finally falls back to +3 points (relaxed). The entire curve lasts approximately 30 minutes, and the general emotional changes are: calm with slight sadness → peaceful and warm → relaxed → joyful → grand and affectionate → pleasant → relaxed.

[0059] The initial music was calm and slightly melancholic, matching the subject's current emotional state and establishing resonance. The music's energy and emotion were then gradually increased, transitioning from peaceful and warm to relaxed, cheerful, and finally grand and soulful, progressively guiding the subject's emotions in a positive direction. The final third of the piece returned to a relaxed level (+3 points), stabilizing the subject's emotions at a moderately positive level and preventing excessive emotional excitement from interfering with subsequent daily activities.

[0060] In a specific example, for subjects with depression, the system provides slow, melancholic music at the beginning of the intervention. As the subjects resonate fully with the music, the emotional tone of the music is gradually changed to help them break free from their depressive state.

[0061] Furthermore, the design of an emotional arousal curve for sleep disorders: Please refer to Figure 3 The emotion curve design unit is configured to design an emotion arousal curve that initially raises the emotion from low to moderate positive levels and gradually decreases to a calm state in the later stages.

[0062] In one specific example, the quantitative change in the mood curve was as follows: from +1 point to +3 points, then gradually decreasing to 0 points (calm). The entire process took approximately 30 minutes. Subjects with sleep disorders used music before falling asleep. The first half of the system provided relaxing music to improve mood, initially raising mood to a moderate level while using positive music to shift focus away from negative thoughts. In the later stages, the emotional and energy levels of the music were gradually reduced to guide relaxation and aid in falling asleep.

[0063] The output of the mood curve design unit is a set of discrete time-mood value correspondence sequences, such as: [(0min, -1 minute), (5min, 0 minutes), (10min, 3 minutes), (15min, 5 minutes), (20min, 7 minutes), (25min, 5 minutes), (30min, 3 minutes)], this sequence data is passed to the music material acquisition module as the matching standard for music selection.

[0064] In this embodiment, the design of the emotional arousal curve is based on the theory of emotional entrainment and the principles of neuromusic therapy. Through the scientific design of emotional change trajectories, it solves the problem of the lack of systematic and targeted emotional regulation in traditional music interventions. Different emotional curves are designed for different types of disorders, achieving precise emotional intervention and significantly improving the intervention effect.

[0065] Furthermore, in this embodiment, the music material acquisition module includes a music material database and a professional music selection interface. It selects music materials that conform to the matching rules from the music material database, taking into account the aesthetic value of the music and its ability to evoke positive visual imagery.

[0066] The music database stores a large number of pre-selected musical works, each with detailed feature tags, including style tags, mood tags, tempo tags, instrument tags, aesthetic ratings, and imagery association ratings. All music materials are instrumental and do not contain lyrics.

[0067] A professional music selection interface is provided for use by music intervention professionals. In a specific example, the left side of the interface displays a mood curve graph, and the right side displays a list of candidate music, which can be filtered by mood, style, instrument, and other criteria. Professionals select music from the database that meets the criteria based on the requirements of each time point in the mood curve.

[0068] The professional music selection interface provides music intervention professionals with a music selection tool to help them filter music that meets their criteria from the database based on emotional curves.

[0069] In a specific example, the music selection is based on the following principles: Emotion matching principle: The emotion label value of the selected music should match the target emotion value of the emotion curve at that time point, with a deviation of ±1 point allowed. For example, for the time point of 7 (cheerful and exciting) in the emotion curve, music with an emotion label of 6-8 can be selected.

[0070] Style preference principle: Prioritize music styles that score higher in group preference tests. For example, if a group preference test shows that subjects prefer classical and new age music, then, provided that emotional matching is satisfied, prioritize selecting music from these two styles.

[0071] Aesthetic Value Principle: Consider the aesthetic value of the music and its ability to evoke positive imagery. Select music with high aesthetic and imagery scores to enhance its appeal and emotional resonance.

[0072] Performance version selection: Different instruments and performers express different emotions for the same piece of music. The music interventionist selects a performance version with appropriate emotional expression of the music and incorporates it into the sequence according to the pre-set music emotional curve.

[0073] In addition, for sleep disorders, the music material acquisition module also includes an audio processing unit, which is configured to technically process the music material to reduce the subject's physiological arousal level and guide relaxation to help fall asleep.

[0074] The audio processing unit is configured to perform at least three of the following technical processes on the music material: 1. Gradually reduce the playback speed of the music file. For example, you can use audio processing software (such as Adobe Audition, Audacity, etc.) to speed up the music file. While keeping the pitch unchanged, reduce the playback speed by 5% to 20%.

[0075] The music material in the later stages (15-30 minutes) of the intervention audio sequence was sped up. For example, music originally at 80 bpm could be gradually reduced to 70 bpm, then 60 bpm. The reduction in music speed caused a rhythmic entrainment effect, which slowed down the subject's heart rate and respiratory rate, thereby reducing physiological arousal levels.

[0076] 2. Simplify the complexity of rhythm and melody, prioritizing musical materials with regular rhythms and simple melodic lines. For music with relatively complex melodies, simpler melodic sections can be extracted using audio editing software.

[0077] 3. Increase the volume transition time between music tracks. During audio splicing, apply a fade-out to the end of each track and a fade-in to the beginning of the next track, with fade-in and fade-out times set to 5-10 seconds. This avoids sudden volume changes during music transitions, making the audio sequence playback smoother and reducing stimulation for the listener. Use logarithmic or S-shaped curves for the fade-out and fade-in curves, rather than non-linear curves, to achieve a more natural transition effect.

[0078] 4. Amplitude limiting processing can be performed. An audio compressor or limiter can be used to compress the dynamic range of the music material, limiting the volume peaks below a set threshold. This avoids sudden volume peaks in the music (such as percussion accents or orchestral overtones) from stimulating and arousing the subject, thus maintaining the stability of the music intensity.

[0079] 5. Prioritize the use of music with regular repetition. Regular, repetitive music patterns reduce novelty and attention capture, facilitating relaxation in subjects. Simultaneously, the repetitive rhythm pattern enhances the physiological entrainment effect, more effectively guiding a decrease in heart rate and respiratory rate.

[0080] The music material acquisition module receives mood curve data from the matching rule setting module. The system also provides a selection interface for filtering music that meets the conditions from the music material database based on the mood curve. Music for sleep disorders is also technically processed by the audio processing unit. Finally, the system passes the selected music material list and audio files to the sequence arrangement module.

[0081] In this embodiment, the system provides a music selection interface based on emotion curves, aesthetic value, and image association characteristics, and provides an audio processing unit with specialized technology for sleep disorders. This provides a structural basis for solving the problems of arbitrary and untargeted music material selection, and achieves a high degree of matching between the emotional expressiveness of the music material and the intervention target, thereby improving the effectiveness of the intervention.

[0082] Furthermore, the sequence arrangement module is used to combine and arrange the musical materials to form an intervention audio sequence. In the emotion curve mode, the core principle of sequence arrangement is to arrange the musical materials according to the chronological order of the emotion arousal curve.

[0083] In a specific example, the sequence orchestration module outputs the complete intervention audio file and metadata (including total duration, music list, mood curve, etc.), uploads it to the cloud storage server, and passes the audio information (file path, download link, etc.) to the playback control module.

[0084] The playback control module is used to periodically provide the intervention audio sequences to the subjects within a preset intervention period. In the emotion curve mode, multiple sets of intervention audio sequences are pre-made using a fixed set approach and are updated regularly within the intervention period.

[0085] In a specific example, for anxiety or depression disorders, a reminder notification is sent before 8 PM daily. At this time, subjects have usually completed their daytime activities and are in a relatively relaxed state, making it easier for them to engage with the emotional experience of the music; at the same time, the improvement in positive emotions helps to improve the subjects' evening mood, preparing them for sleep at night.

[0086] In a specific example, for sleep disorders, a reminder is sent 30-60 minutes before bedtime as set by the participant. Participants can customize their bedtime in the settings, and the system automatically calculates the reminder time based on their bedtime. The reason for choosing the 30-60 minute bedtime period is that the relaxing effect of music can directly affect the sleep process, reduce physiological arousal levels, and shorten the sleep latency.

[0087] Example 3 In this embodiment, when the matching rule setting module adopts a direct preference mapping strategy, the specific implementation methods of each module in the system are as follows. This implementation mode places greater emphasis on individualization and dynamism, and can respond to the current needs of the subjects in real time.

[0088] In this embodiment, the music preference acquisition module is configured to repeatedly collect the subject's current music preference information at multiple time points within the preset intervention period. The music preference information includes at least three dimensions of music emotion, music style, music intensity, instrument type, music tempo, and degree of music variation, preferably including all six dimensions.

[0089] Unlike Example 2, which used a one-time group test, this example uses a dynamic and personalized preference collection method. For example, before each week's intervention, participants' current music preferences are collected through questionnaires.

[0090] In a specific example, the design of the six dimensions includes: Musical Emotion Dimension: Based on the results of interviews with participants after they listened to music, the classification of musical emotions was formed by organizing and categorizing participants' authentic descriptions of the music. This classification method, based on the participants' genuine emotional experiences and verbal expressions, is closer to the participants' actual feelings than academic emotion classifications, making it easier for participants to accurately express their preferences.

[0091] Music style dimension: This dimension covers the most clinically preferred music styles while retaining open options to accommodate individual differences.

[0092] Music intensity dimension: Music intensity affects the physiological arousal level of the subjects. Different emotional states and intervention stages require different music intensities.

[0093] Instrument type dimension: Different instruments have different timbre characteristics and emotional expressiveness. Choosing the instrument preferred by the test takers can enhance the appeal and resonance of the music.

[0094] Music tempo dimension: Music tempo directly affects the rhythm carry-over effect, thereby affecting heart rate and respiratory rate.

[0095] Musical Variation Dimension: The degree of variation in music affects the subjects' level of concentration and emotional arousal. Simple, repetitive music is more suitable for relaxation and sleep aid, while rich and varied music is more likely to attract attention and enhance mood.

[0096] The system backend receives the subjects' preference information in real time, stores it in the database, and associates it with the subject's ID, date, and week.

[0097] In a specific example, the data table structure may include: subject ID, date and time of completion, week, music mood selection, music style selection, music intensity selection, instrument type selection, music tempo selection, and music variation level selection.

[0098] This embodiment employs an individualized and dynamic preference collection method, which can reflect the subject's current music needs and emotional state in real time, solving the problem that group testing cannot reflect individual differences and dynamic changes. Multi-dimensional preference collection comprehensively covers the key elements affecting the music experience, providing a sufficient information foundation for precise music customization.

[0099] In this embodiment, the matching rule setting module is configured to directly map the features of each dimension in the music preference information to the corresponding feature requirements of the music material.

[0100] Unlike the indirect matching based on emotion curves in Example 2, this example uses a direct mapping method to directly convert the preferences expressed by the subjects into the feature requirements of the music material.

[0101] Studies have shown that listening to unfamiliar but highly enjoyable music can activate the bilateral limbic and paralimbic areas, generating positive emotional experiences. William et al.'s research also found that people's physiological and psychological responses to favorite music showed a decrease in anxiety and an increase in relaxation from before to after the test.

[0102] In this embodiment, the matching rule setting module includes a preference information parsing subunit, a feature mapping subunit, and a description statement generation subunit.

[0103] Specifically, the preference information parsing subunit reads the latest preference questionnaire data submitted by the subjects from the database, parses the content of each field, and identifies the subjects' choices in each dimension.

[0104] The feature mapping subunit maps questionnaire options to musical feature descriptions according to preset mapping rules. It maps six dimensions of preference information to musical feature requirements: musical mood → emotional expressiveness of music; musical style → genre of music; musical intensity → dynamic markings of music; instrument type → arrangement and orchestration of music; musical tempo → tempo markings of music; and degree of musical variation → complexity and variability of music.

[0105] The description statement generation subunit concatenates feature descriptions from each dimension into a complete natural language description statement.

[0106] In a specific example, if a participant's questionnaire answer is: the music's mood is calm and soothing, the music style is Western classical music, the music intensity is soft, the instruments are flute, piano, and cello, the music tempo is slow, and the degree of musical variation is simple repetition, then it maps to the music description statement: "Soft, slow, simple repetition, calm and soothing Western classical music, played on flute, piano, and cello."

[0107] In summary, the matching rule setting module outputs a music description statement, which is then passed to the music material acquisition module as input for music generation or retrieval.

[0108] In this embodiment, the direct mapping matching rules fully respect the individualized needs of the subjects, solving the problem that standardized intervention programs cannot meet individual differences. By transforming subjective preferences into objective musical feature requirements, clear goals are provided for subsequent music generation or retrieval, achieving subject-centered individualized intervention.

[0109] In this embodiment, the music material acquisition module includes: a preference conversion unit, a music matching unit, a rating collection unit, and an intelligent filtering unit.

[0110] The preference conversion unit is used to convert the music preference information into music description statements, receive the music description statements from the matching rule setting module, and perform formatting processing.

[0111] The music matching unit is used to generate or retrieve multiple candidate music materials based on the music description statement.

[0112] The music matching unit can be implemented using one or a combination of the following two methods: Method 1: Music generation based on artificial intelligence The music matching unit includes an AI music generation subunit, which is configured to: call the AI ​​music generation platform through an API interface; generate multiple candidate music materials; and store the generated music files.

[0113] Method 2: Intelligent search based on music database Key features are extracted from the musical descriptions, such as style, instrument, mood, tempo, and degree of variation. Multi-condition searches are performed in the music database based on these features. A matching score is calculated for each retrieved piece of music. The music is then sorted by matching score, and the top 10-20 pieces with the highest scores are selected as candidates.

[0114] Method 3: Hybrid Method Combining the two methods above, the music database is first searched. If a sufficient number (e.g., more than 10) of highly matched music (match score > 35) are found, they are used directly. If the search results are insufficient or the match is not high, the AI ​​music generation system is called to supplement the generation.

[0115] Furthermore, the scoring unit is used to collect the subjects' preference ratings for the candidate music materials.

[0116] The intelligent filtering unit is used to filter music materials with higher scores based on the preference rating.

[0117] In summary, the music material acquisition module passes the selected music data to the sequence arrangement module.

[0118] Furthermore, the sequence arrangement module is used to combine and arrange the musical materials to form an intervention audio sequence. In the direct mapping mode, the core principle of sequence arrangement is to combine them based on the subject's preference rating.

[0119] The playback control module is used to periodically provide the intervention audio sequence to the subjects within a preset intervention period. In direct mapping mode, a dynamic generation method is used, and the audio is regenerated weekly based on the subjects' latest preferences.

[0120] Example 4 This invention provides a personalized music intervention method; please refer to [the relevant documentation]. Figure 4 This includes the following steps: S1. Obtain the subject's music preference information; S2. Determine the matching rules between music preferences and intervention needs based on the intervention goals; S3. Based on the music preference information and the matching rules, obtain multiple music materials that meet the requirements; S4. Combine and arrange the music materials to form an intervention audio sequence.

[0121] Furthermore, in response to the different matching rules in step S2 of this individualized music intervention method, this embodiment provides two specific music intervention methods.

[0122] First, the matching rules in step S2 include: Based on the theory of emotion entrainment, a pre-designed emotion arousal curve is created, which defines the trajectory of changes in musical emotion during the intervention period. The emotional expressiveness of the musical material must match the corresponding time point of the emotional arousal curve.

[0123] Based on the first matching rule described above, this embodiment of the invention provides a personalized music intervention method based on an emotional arousal curve. The method specifically includes the following steps: S11: Obtain the subject's music preference information.

[0124] In this embodiment, the music preference information is obtained by conducting a music style preference scale test on the target group. Specifically, the STOMPS-R (Short Test of Music Preferences-Revised) scale can be used to conduct a group test on the subjects to understand their basic music preferences.

[0125] S12: Determine the matching rules between music preferences and intervention needs based on the intervention goals.

[0126] In this embodiment, the matching rule is based on the emotion entrainment theory and a preset emotion arousal curve is designed. The emotion arousal curve specifies the trajectory of music emotion changes during the intervention period and requires that the emotional expressiveness of the music material matches the corresponding time point of the emotion arousal curve.

[0127] Emotion curve design for different intervention goals: (1) Please continue to refer to Figure 2 The design of emotional arousal curves for anxiety or depressive disorders specifically includes: The design features of the emotional arousal curve are: starting from a negative or low positive emotional starting point, gradually rising to a positive emotional peak, and then falling back to a moderate positive emotional level.

[0128] In a specific example, within the emotional quantification scoring system, the emotional arousal curve changes as follows: starting from -1 point (calm with slight sadness), it gradually increases to +7 points (joyful, grand, and affectionate), and finally falls back to +3 points (relaxed). The entire curve lasts approximately 30 minutes, with the emotional changes roughly as follows: calm with slight sadness → peaceful and warm → relaxed → joyful → grand and affectionate → pleasant → relaxed.

[0129] (2) Please continue to refer to Figure 3 The design of emotional arousal curves for sleep disorders specifically includes: The design features of the emotional arousal curve are: initially, it raises the emotional level from low to moderately positive, and then gradually decreases to a calm state in the later stages.

[0130] In one specific example, the quantitative change in the emotion curve was as follows: from +1 point to +3 points, and then gradually decreasing to 0 points (calm). The entire process took approximately 30 minutes.

[0131] Specifically, the first half uses relaxing music, with the music design initially raising the mood to a moderate level (1→3→0), and gradually lowering the mood and energy level of the music in the middle and later stages.

[0132] In this embodiment, the design of the emotional arousal curve is based on the theory of emotional entrainment and the principles of neuromusic therapy. Through the scientific design of emotional change trajectories, it solves the problem of the lack of systematic and targeted emotional regulation in traditional music interventions. Different emotional curves are designed for different goals, achieving precise and personalized design.

[0133] S13: Based on the music preference information and the matching rules, obtain multiple music materials that meet the requirements.

[0134] In this embodiment, music materials that meet the matching rules are selected from a music library. During the selection process, not only the match between the music and the emotional curve are considered, but also the aesthetic value of the music and its ability to evoke positive imagery.

[0135] Furthermore, it also includes technical processing of the music material, the technical processing including at least three of the following: Gradually reduce the playback speed of music: You can use speed adjustment to gradually slow down the music.

[0136] Simplify the complexity of rhythm and melody: The rhythm and melody should be simple and stable. Music with strong regularity and repetition can be selected to reduce the fluctuations and changes in the music.

[0137] Increase the volume transition time between music tracks: In the audio processing section, the volume transition between tracks can be increased.

[0138] Amplitude limiting can be applied to avoid excessive stimulation and arousal, thus maintaining the stability of the music intensity.

[0139] Prioritize music materials with regular repetition characteristics.

[0140] In this embodiment, music selection is based on emotion curves, aesthetic value, and image association characteristics, which solves the problems of randomness and lack of specificity in music material selection and achieves a high degree of matching between the emotional expressiveness of the music material and the intervention target.

[0141] S14: Combine and arrange the music materials to form an intervention audio sequence.

[0142] The music materials are arranged in chronological order according to the emotional arousal curve. Preferably, 6 to 10 music segments are combined to form an intervention audio sequence with a duration of 20 to 40 minutes.

[0143] In a specific example, a 30-minute intervention audio sequence for anxiety or depressive disorders may include: 0-3 minutes: Calm and slightly sad music (emotional score -1 point); 3-8 minutes: Peaceful and warm music (emotional score 0-2 points); 8–12 minutes: Relaxing music (emotional score 3–4 points); 12–17 minutes: Upbeat music (emotional score 5–6); 17-22 minutes: Magnificent and soulful music (emotional score 7 points); 22-26 minutes: Pleasant music (emotional score 5-6); 26-30 minutes: Relaxing music (emotional score 3 points).

[0144] In another specific example, a 30-minute intervention audio sequence for sleep disorders includes: 0-10 minutes: Relaxing music, gradually raising the mood from 1 point to 3 points; 10-20 minutes: Calm music, gradually lowering the mood from 3 to 1; 20-30 minutes: Very calm music, the mood gradually decreases from 1 point to 0 points, the music speed becomes slower and slower, and the repetition becomes stronger and stronger.

[0145] In one specific example, multiple sets of intervention audio sequences are pre-made and updated every 5 to 10 days within the preset intervention period.

[0146] Furthermore, based on the different matching rules in step S2, this embodiment also provides another specific individualized music intervention method that meets individual needs. The individualized music intervention method that meets individual needs differs from the detailed steps of the matching rules and music material acquisition method of the aforementioned individualized music intervention method based on the emotional arousal curve.

[0147] Specifically, in this embodiment, the individualized music intervention method that meets individual needs includes the following steps: S21: Obtain the subject's music preference information.

[0148] In this embodiment, the step of obtaining the subject's music preference information is repeatedly executed at multiple time points within the preset intervention period. Each time, the subject's current music preference information is collected. The music preference information includes at least three dimensions of music emotion, music style, music intensity, instrument type, music tempo, and degree of music variation, preferably including all six dimensions.

[0149] Furthermore, this embodiment employs a dynamic and personalized preference collection method. Within a predetermined timeframe, such as weekly, the subject's current music preferences are collected via questionnaire.

[0150] Dynamic data acquisition can be implemented in the following ways: In a specific example, a music preference questionnaire can be pushed to participants via an application, and participants can complete the questionnaire on their mobile devices. The system backend receives and stores the participants' preference information in real time, which serves as the basis for generating music for the week.

[0151] S22: Determine the matching rules between music preferences and intervention needs based on the intervention goals.

[0152] In this embodiment, the matching rule in step S2 includes: directly mapping the features of each dimension in the music preference information to the corresponding feature requirements of the music material. This directly transforms the preferences expressed by the subject into feature requirements of the music material.

[0153] Furthermore, the establishment of mapping rules includes the following process: Mapping the six dimensions of preference information to music feature requirements: Musical mood → The emotional expressiveness of music; Music style → Genre of music; Music intensity → The dynamic markings of music; Instrument type → Music arrangement and orchestration; Music tempo → The tempo markings for the music; The degree of musical variation → the complexity and variability of music.

[0154] In a specific example, if the subject's questionnaire answers are: Musical mood: calm and soothing; Musical style: Western classical music; Music intensity: weak; Musical instruments: flute, piano, cello; Music tempo: Slow; Musical variation: simple and repetitive.

[0155] This is mapped to the following musical characteristics: soft, slow, simple, repetitive, calm and soothing Western classical music, played on flute, piano, and cello.

[0156] In this embodiment, the direct mapping matching rules fully respect the individualized needs of the subjects, solving the problem that standardized intervention programs cannot meet individual differences. By transforming subjective preferences into objective musical feature requirements, a clear target is provided for subsequent music generation or retrieval.

[0157] S23: Based on the music preference information and the matching rules, obtain multiple music materials that meet the requirements.

[0158] In this embodiment, please refer to Figure 5 The steps for obtaining multiple suitable music materials specifically include: S231: Convert the music preference information into a music description statement.

[0159] In a specific example, the backend system can use a RESTful interface to receive and structure music preference data, and then concatenate the questionnaire results completed by the subjects in the application to form a string of music feature descriptions in natural language.

[0160] For example, the subject's questionnaire response was: Musical mood: calm and soothing; Musical style: Western classical music; Music intensity: weak; Musical instruments: flute, piano, cello; Music tempo: Slow; Musical variation: simple and repetitive.

[0161] The music description is compiled using a RESTful interface: soft, slow, simple, repetitive, calm and soothing Western classical music played on flute, piano, and cello. This natural language description format facilitates subsequent processing by an AI-powered music generation system.

[0162] S232: Generate or retrieve multiple candidate music materials from the music description statement.

[0163] Step S232 can be implemented in one or a combination of the following two ways: 1. Music generation based on artificial intelligence.

[0164] For example, you can call the API provided by the music AI generation platform, and based on the weekly music preference descriptions of the subjects organized by the RESTful interface, select the music generation V4 model and pure instrumental mode to initiate a task to generate 15 personalized music tracks.

[0165] 2. Intelligent retrieval based on music database.

[0166] Using key features from the music description as search criteria, an intelligent search is performed in a pre-established music material database to filter out music works that meet the criteria.

[0167] 3. Mixing method Combining the two methods mentioned above, the music database is first searched. If a sufficient number of highly matching songs are found, they are used directly. If the search results are insufficient or the matching degree is not high, the artificial intelligence music generation system is called to supplement the generation.

[0168] S233: Subjects rate the candidate music materials based on their preferences.

[0169] S234: Filter music materials with higher scores based on the preference ratings.

[0170] In a specific example, the processes S231 to S234 described above may include: (1) The music preference information was collected through a digital questionnaire: the subjects filled out an electronic questionnaire and the data was transmitted to the back-end server in real time.

[0171] (2) The music preference information is structured through the backend interface: After receiving the questionnaire data, the backend server converts the selection of each dimension into music description statements through a preset algorithm.

[0172] (3) Automatically call the API interface of the AI ​​music generation system: The backend system automatically sends a generation request to the API of the music AI generation platform without manual operation.

[0173] (4) Store the generated candidate music materials to the cloud: The generated music files are automatically downloaded and stored in the specified directory of the cloud server.

[0174] (5) Automatically push music generation completion notification to the subject's terminal: Push a message notification to the subject to inform them that the music has been generated and they can listen to and score it.

[0175] The system code for the above process is as follows: <?php function aiAudioGenerate($prompt,$access_token,$patient_id) { / / $prompt is a concatenated string representing the personalized music style preference settings completed by the test subject on the mini-program. / / $prompt = 'Soft, slow, simple, repetitive, calm and soothing Western classical music played on flute, piano, and cello' $curl = curl_init(); $postData1 = array( "prompt" => $prompt, "customMode" => false, "instrumental" => true, "model" =>"V4", "callBackUrl" => "https: / / XXX.YYY.ZZZ / version0008 / public / xcxEvaluation / getAIAudioCallback" / / Actual callback address ); / / Disable SSL verification curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false); curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, false); / / Call the music AI generation platform interface curl_setopt_array($curl, array( CURLOPT_URL =>'https: / / apibox.erweima.ai / api / v1 / generate', CURLOPT_RETURNTRANSFER =>true, CURLOPT_ENCODING =>'', CURLOPT_MAXREDIRS => 10, CURLOPT_TIMEOUT => 10000, / / Adjust timeout CURLOPT_FOLLOWLOCATION =>true, CURLOPT_HTTP_VERSION =>CURL_HTTP_VERSION_1_1, CURLOPT_CUSTOMREQUEST =>'POST', CURLOPT_POSTFIELDS => json_encode($postData1), / / Use json_encode to generate a JSON string CURLOPT_HTTPHEADER => array( 'Content-Type: application / json', 'Accept: application / json', 'Authorization: Bearer ' . $access_token ), )); $arrayData = []; $response = curl_exec($curl); / / Generate a task ID and persistently store it in a cloud database. $returnArr['response'] = $arrayData; curl_close($curl); return $returnArr; } The entire process is highly automated, which greatly reduces labor costs and improves processing efficiency.

[0176] S24: Combine and arrange the music materials to form an intervention audio sequence.

[0177] In this embodiment, every 5 to 10 days during the intervention period, preferably weekly, the steps of obtaining music preference information (S21), determining matching rules (S22), obtaining music materials (S23), and combining and arranging (S24) are re-executed to dynamically generate the intervention audio sequence used during that time period.

[0178] In addition, the personalized music intervention method provided by the present invention also includes the step of: digitizing the music elements of the integrated audio.

[0179] In this embodiment, data analysis is performed on the musical elements of the final audio material, such as tempo, style, and loudness, to obtain digital audio preference data for each subject, which will serve as the basis for future data analysis of the subject. Simultaneously, the subject's musical data characteristics and a music preference database are formed.

[0180] In summary, the personalized music intervention system and method of the present invention have the following beneficial effects: By acquiring participants' music preferences and matching them with appropriate music materials, the appeal of the music and participant engagement were enhanced. An emotional arousal curve was designed using the theory of emotional entrainment, addressing the lack of systematicity and specificity in traditional music interventions. Differentiated emotional curves and technical solutions were designed for different disorder types. Furthermore, artificial intelligence music generation technology was introduced to automatically generate personalized music based on participants' preferences, improving the efficiency of personalized music acquisition. This provides a comprehensive intervention system for achieving personalized, scientific, intelligent, and scalable systematic intervention methods.

[0181] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A personalized music intervention system, characterized in that, include: The music preference acquisition module is used to obtain the music preference information of the subjects; The matching rule setting module is used to determine the matching rules between music preferences and intervention needs based on the intervention objectives; The music material acquisition module is used to acquire multiple music materials that meet the requirements based on the music preference information and the matching rules. The sequence arrangement module is used to combine and arrange the music materials to form an intervention audio sequence; The playback control module is used to periodically provide the intervention audio sequence to the subject within a preset intervention period.

2. The personalized music intervention system according to claim 1, characterized in that: The matching rule setting module includes an emotion curve design unit, which is used to design an emotion arousal curve based on the emotion entrainment theory. The emotion arousal curve specifies the trajectory of changes in musical emotion during the intervention period and requires that the emotional expressiveness of the musical material matches the corresponding time point of the emotion arousal curve.

3. The personalized music intervention system according to claim 2, characterized in that: The emotion curve design unit is configured as follows: For anxiety or depression disorders, an emotional arousal curve is designed that starts from a negative or low positive emotional level, gradually rises to a peak of positive emotion, and then falls back to a moderate level of positive emotion. For sleep disorders, an emotional arousal curve was designed, which initially raises positive emotions to a moderate level and then gradually decreases them to a calm state in the later stages.

4. The personalized music intervention system according to claim 3, characterized in that: The music material acquisition module also includes an audio processing unit, which, for sleep disorders, is configured to perform at least three technical processes on the music material: Gradually reduce the playback speed of the music; Simplify the complexity of rhythm and melody; Increase the volume transition time between music tracks; Amplitude limiting measures were implemented. Prioritize music materials with regular repetition characteristics.

5. The personalized music intervention system according to claim 2, characterized in that: The music preference collection module is configured to test the target group using a music style preference scale and select music style types with higher scores as preference references. The music material acquisition module includes a music material database and a professional music selection interface. It selects music materials that meet the matching rules from the music material database, taking into account the aesthetic value of the music and its ability to evoke positive imagery.

6. The personalized music intervention system according to claim 1, characterized in that: The matching rule setting module is configured to directly map the features of each dimension in the music preference information to the corresponding feature requirements of the music material.

7. The personalized music intervention system according to claim 6, characterized in that: The music preference acquisition module is configured to repeatedly collect the subject's current music preference information at multiple time points within the preset intervention period. The music preference information includes at least three dimensions: music emotion, music style, music intensity, instrument type, music tempo, and degree of music variation.

8. The personalized music intervention system according to claim 6, characterized in that: The music material acquisition module includes: A preference conversion unit is used to convert the music preference information into music description statements; A music matching unit is used to generate or retrieve multiple candidate music materials based on the music description statement; The scoring collection unit is used to collect the subjects' preference ratings for the candidate music materials; The intelligent filtering unit is used to filter music materials with higher scores based on the preference rating.

9. The personalized music intervention system according to claim 8, characterized in that: The music matching unit includes an artificial intelligence music generation subunit and / or a music database retrieval subunit; The AI ​​music generation subunit is configured as follows: Call the AI-powered music generation platform via API interface; Generate multiple candidate music materials; and Store the generated music file; The music database retrieval subunit is configured as follows: Perform a multi-condition search in the music material database based on the aforementioned music description statement; Calculate the match score of the retrieved music; as well as Sort by matching score and select multiple candidate music materials.

10. A personalized music intervention method, characterized in that, include: Obtain the subjects' music preference information; Determine the matching rules between music preferences and intervention needs based on the intervention goals; Based on the music preference information and the matching rules, multiple music materials that meet the requirements are obtained; as well as The music materials are combined and arranged to form an interventional audio sequence; The matching rules include: Based on the theory of emotional entrainment, a preset emotional arousal curve is designed. The emotional arousal curve defines the trajectory of changes in musical emotions during the intervention period, requiring that the emotional expressiveness of the musical material matches the corresponding time point of the emotional arousal curve. or The features of each dimension in the music preference information are directly mapped to the corresponding feature requirements of the music material.

Images