Integrated intelligent audio-video playing system

Abstract

The invention relates to the technical field of audio-video intelligent optimization, and discloses an integrated intelligent audio-video playing system, which comprises a content acquisition module, a user analysis module, a playing control module, an energy-saving optimization module, a state monitoring module, a cloud synchronization module and an emergency processing module, by setting an intelligent sensing module, when video content is played adaptively, personalized matching of video recommendation is ensured by formulating double analysis standards of user emotion and environmental parameters and setting differentiated recommendation thresholds for different scenes, and meanwhile, environmental noise and user physiological sensor data are collected in real time; therefore, interference factors and instant emotional states of the environment where the user is located can be dynamically perceived, the problem of recommendation dislocation caused by environment interference is eliminated, high conformity of played content and user requirements is guaranteed, and the recommendation error rate is reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent audio-visual optimization technology, specifically to an integrated intelligent audio-visual playback system. Background Technology

[0002] With social progress, improved living standards, and advancements in science and technology, audio-visual equipment is increasingly used for people's entertainment and leisure, enhancing their quality of life. The demand for intelligent audio-visual playback systems in mobile devices is growing daily. Audio-visual equipment can play movies and songs, providing entertainment for people. With the rapid development of high technology in various industries and the integration of the country with the world, people's horizons have been further broadened compared to the past. People are no longer content with monotonous and outdated entertainment programs, and rigid performance modes have been eliminated.

[0003] Currently, due to the diverse needs of audio-visual playback scenarios, the recommendation algorithms used for adaptive playback of audio-visual content are based on users' historical behavior data and cannot perceive users' current emotional state and environmental interference factors in real time. When users are in noisy environments or emotionally fluctuating states, the recommended content may be misaligned with the user's immediate needs, failing to guarantee accurate matching of playback content. Furthermore, during multi-device collaborative playback, the inability to dynamically detect network bandwidth fluctuations and device performance differences can lead to audio-visual asynchrony and stuttering during cross-device playback, and the lack of adaptive error correction mechanisms in case of playback abnormalities. Regarding energy efficiency optimization, the inconsistent hardware configurations of playback devices prevent a fine-grained balance between playback quality and energy consumption during dynamic energy-saving adjustments, resulting in system overheating under high load and energy waste under low load, affecting the smoothness of audio-visual playback and system battery life.

[0004] Therefore, an integrated intelligent audio-visual playback system is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an integrated intelligent audio-visual playback system, which solves the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an integrated intelligent audio-visual playback system, the system comprising: The content acquisition module acquires multi-source audio and video content data from local storage devices, network streaming media services, and cloud databases. Its built-in decoder is compatible with high-definition video streams, 4K video streams, and compressed audio streams, and supports real-time data capture and conversion. The user analysis module, based on deep learning models and natural language processing algorithms, analyzes voice commands, gesture recognition, and touch operation data input from external devices, and generates preference analysis data containing user age groups, interest tags, and viewing frequency parameters through pattern recognition. The playback control module dynamically adjusts the playback order, volume level, and picture quality settings based on this preference analysis data to achieve adaptive playback control. The energy-saving optimization module collects light intensity, ambient temperature and system load parameters through sensors, and uses dynamic bitrate adjustment technology to optimize the power consumption of the processor and display device, reducing screen brightness in low light environment and reducing processor frequency under high temperature conditions. The status monitoring module monitors CPU utilization, memory usage, and network bandwidth parameters in real time. It triggers three levels of alerts through anomaly detection algorithms: Level 1 alerts optimize resource allocation, Level 2 alerts initiate data cleanup, and Level 3 alerts generate a forced hibernation command.

[0007] Preferably, the system further includes a cloud synchronization module and an emergency response module: The cloud synchronization module establishes an encrypted transmission channel to synchronize playback progress and user preference data to the cloud server in real time, supporting cross-device retrieval and resume playback; The emergency handling module has an embedded fault prediction algorithm. When the system performance reports an abnormality, it automatically switches to the backup playback mode and maintains basic image quality output until the fault is resolved, ensuring playback continuity.

[0008] Preferably, the system includes the following steps: S1. Obtain audio-visual content data and user input data; S2. Based on the audio-visual content data and the user input data, perform user preference analysis processing to generate user preference analysis data; S3. Based on the user preference analysis data and the audio-visual content data, perform intelligent playback control processing to generate playback control instruction data; S4. Based on the playback control command data and environmental parameter data, perform energy-saving optimization processing to generate energy-saving strategy data; S5. Perform audio-visual playback execution processing based on the energy-saving strategy data and the playback control instruction data to generate audio-visual playback output data; S6. Based on the audio-visual playback output data and user feedback data, perform real-time adjustment processing to generate playback adjustment instruction data; S7. Construct summary data for audio and video playback, perform system status monitoring and processing, and generate system status report data.

[0009] Preferably, step S1 includes the following steps: S11. Extract uncompressed audio and video raw data from the local storage device through the content acquisition module, and generate video data V and audio data A. The video data V represents the combination parameters of frame sequence parameters and resolution parameters established with the time axis as a one-dimensional coordinate axis. The video data includes high-definition video streams and 4K video streams. S12. Capture user interaction data from external devices through the user input module and generate user input data U. The user input data U includes voice command data, gesture recognition data and touch operation data. The external devices include microphone, camera and touch screen.

[0010] Preferably, step S2 includes the following steps: S21. Import the generated video data V, audio data A, and user input data U into the intelligent analysis platform, and use a deep learning algorithm to search for target user behavior feature information in the intelligent analysis platform according to preference keywords, and generate user preference analysis data P, where the unit of P is preference weight value; S22. Based on the user preference analysis data P, the K-means clustering algorithm is used to classify user groups. Euclidean distance is used as the metric to calculate the sample similarity in the feature space, generating user group classification data; wherein: I. Define the initialization rule for cluster centers: Randomly select k sample points from the preference analysis data P as the initial cluster centers. ; II. Perform Min-Max normalization on each feature dimension of the user preference analysis data P, so that all feature values ​​are scaled to the [0,1] interval: in Let m be the minimum value of the m-th feature among all samples. Let m be the maximum value of the m-th feature in all samples. For user j, the original measurement value on feature m. These are the normalized user feature values. Let m be the original center value of cluster i on feature m. These are the normalized cluster center eigenvalues; III. Calculate the Euclidean distance metric: For each user sample point Calculate the distance from each normalized cluster center: in The value is the Euclidean distance. Let j be the feature vector of the j-th user. Let be the feature vector of the i-th cluster center, n be the dimension of the feature space, and m be the feature dimension index (m = 1, 2, ..., n). Let m be the normalized feature value of user j in the m-th dimension. Let be the normalized center value of the m-th dimension of cluster i; IV. Assigning samples to nearest neighbor clusters: Assigning sample points to nearest neighbor clusters. Assign to the category of the cluster center with the smallest distance; V. Update cluster center locations: Recalculate the center point of each cluster: in For the i-th cluster set, Let i be the new position of the i-th cluster center after the (t+1)th iteration, where i is the cluster center number (i = 1, 2, ..., k). This serves as an identifier for the number of iterations. VI. Iterate through the steps until the cluster center positions converge, and the offset between adjacent iteration centers is less than the threshold. The final user group classification data is output, including age group classification labels, interest clustering identifiers, and high-frequency behavioral feature codes.

[0011] Preferably, step S3 includes the following steps: S31. Obtain the user preference analysis data P and the audio-visual content data; S32. Match the user preference analysis data P with the pre-stored audio-visual content library, and generate playback control command data based on the matching results. ; When a match is successful, an adaptive playback command is output; If a match fails, the default playback command is output.

[0012] Preferably, step S4 includes the following steps: S41. When the environmental parameter data indicates a high-energy-consumption mode, a dynamic bitrate adjustment algorithm is used to adjust the playback control command data. The resolution parameters are adjusted downward according to the energy-saving threshold, and energy-saving strategy data E is generated. S42. Based on the energy-saving strategy data E, perform power consumption prediction processing to generate expected energy-saving rate data, which represents the percentage of energy savings per unit time.

[0013] Preferably, step S5 includes the following steps: S51. The generated energy-saving strategy data E and the playback control command data are... The corresponding control parameters are arranged aligned according to the playback timeline, and the audio-visual playback output data is constructed. : in Output data for audio and video playback. For energy-saving strategy data, For playback control command data; S52, Based on the audio / video playback output data The driver performs audio-visual rendering processing on the playback device to generate real-time playback stream data.

[0014] Preferably, step S6 includes the following steps: S61. Establish a standard adjustment strategy matrix corresponding to different user feedback types: in Adjust the strategy matrix to the standard. This represents the adjustment strategy corresponding to the k-th feedback type, including pause strategy, fast forward strategy, and volume adjustment strategy. ; S62, Combine the playback adjustment command data with the... In Perform strategy matching to generate optimal adjustment instruction data.

[0015] Preferably, step S7 includes the following steps: S71, output the audio / video playback data The playback adjustment command data and the system status report data are combined to construct a summary of audio and video playback data. : in Data is compiled for video and audio playback. For system status identification; S72. The intelligent analysis platform calls the monitoring program to perform real-time performance analysis on R based on the system status report data, and generates system health report data.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. In this invention, by setting up an intelligent sensing module, when adaptively playing audio-visual content, it establishes dual analysis standards for user emotions and environmental parameters, and sets differentiated recommendation thresholds for different scenarios to ensure personalized matching of audio-visual recommendations. At the same time, it collects environmental noise and user physiological sensor data in real time, which can dynamically sense the interference factors in the user's environment and the user's immediate emotional state, eliminate the recommendation misalignment problem caused by environmental interference, ensure a high degree of fit between the playback content and the user's needs, and reduce the recommendation error rate.

[0017] 2. By setting up a collaborative control module, when multiple devices are playing together, the system dynamically adjusts the audio and video stream synchronization parameters by calculating the network bandwidth fluctuation value and the device performance difference index in real time. This enables the system to automatically compensate for cross-device transmission delays. Furthermore, when a risk of playback interruption is detected, a distributed buffer retransmission mechanism is immediately activated. By preloading key data frames, the audio and video deviations are corrected in real time, ensuring smooth playback even under conditions of device performance differences and network fluctuations, and eliminating audio and video desynchronization and stuttering.

[0018] 3. In this invention, by setting up a dynamic balancing module, when optimizing playback energy consumption, a quantitative mapping model between image quality level and power consumption threshold is established. High-load areas and low-load areas are segmented in real time for graded control. Based on hardware temperature and battery capacity data under different scenarios, the optimal energy efficiency solution is dynamically matched, enabling the system to seamlessly switch between 4K high-definition playback and energy-saving mode. This solves the problem of energy efficiency imbalance in traditional systems and avoids overheating under high load and energy waste under low load. Attached Figure Description

[0019] Figure 1 This is a structural diagram of an integrated intelligent audio-visual playback system according to the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 This integrated intelligent audio-visual playback system includes: The content acquisition module acquires multi-source audio and video content data from local storage devices, network streaming media services, and cloud databases. Its built-in decoder is compatible with high-definition video streams, 4K video streams, and compressed audio streams, and supports real-time data capture and conversion. The user analysis module, based on deep learning models and natural language processing algorithms, analyzes voice commands, gesture recognition, and touch operation data input from external devices, and generates preference analysis data containing user age groups, interest tags, and viewing frequency parameters through pattern recognition. The playback control module dynamically adjusts the playback order, volume level, and picture quality settings based on this preference analysis data to achieve adaptive playback control. The energy-saving optimization module collects light intensity, ambient temperature and system load parameters through sensors, and uses dynamic bitrate adjustment technology to optimize the power consumption of the processor and display device, reducing screen brightness in low light environment and reducing processor frequency under high temperature conditions. The status monitoring module monitors CPU utilization, memory usage, and network bandwidth parameters in real time. It triggers three levels of alerts through anomaly detection algorithms: Level 1 alerts optimize resource allocation, Level 2 alerts initiate data cleanup, and Level 3 alerts generate a forced hibernation command.

[0022] The system also includes a cloud synchronization module and an emergency response module: The cloud synchronization module establishes an encrypted transmission channel to synchronize playback progress and user preference data to the cloud server in real time, supporting cross-device retrieval and resume playback; The emergency handling module has an embedded fault prediction algorithm. When the system performance reports an abnormality, it automatically switches to the backup playback mode and maintains basic image quality output until the fault is resolved, ensuring playback continuity.

[0023] The system includes the following steps: S1. Obtain audio-visual content data and user input data; S2. Based on audio-visual content data and user input data, perform user preference analysis and processing to generate user preference analysis data; S3. Based on user preference analysis data and audio-visual content data, perform intelligent playback control processing to generate playback control command data; S4. Based on playback control command data and environmental parameter data, perform energy-saving optimization processing to generate energy-saving strategy data; S5. Perform audio-visual playback execution processing based on energy-saving strategy data and playback control command data, and generate audio-visual playback output data; S6. Based on audio and video playback output data and user feedback data, perform real-time adjustment processing to generate playback adjustment instruction data; S7. Construct summary data for audio and video playback, perform system status monitoring and processing, and generate system status report data.

[0024] S1 includes the following steps: S11. Extract uncompressed audio and video raw data from the local storage device through the content acquisition module, and generate video data V and audio data A. Video data V represents the combination parameters of frame sequence parameters and resolution parameters established with the time axis as a one-dimensional coordinate axis. The video data includes high-definition video streams and 4K video streams. S12. Capture user interaction data from external devices through the user input module and generate user input data U. User input data U includes voice command data, gesture recognition data and touch operation data. External devices include microphone, camera and touch screen.

[0025] S2 includes the following steps: S21. Import the generated video data V, audio data A, and user input data U into the intelligent analysis platform, and use a deep learning algorithm to search for target user behavior feature information in the intelligent analysis platform according to preference keywords, and generate user preference analysis data P, where the unit of P is preference weight value; S22. Based on user preference analysis data P, K-means clustering algorithm is used to classify user groups. Euclidean distance is used as the metric to calculate sample similarity in the feature space, generating user group classification data; where: I. Define the initialization rule for cluster centers: Randomly select k sample points from the preference analysis data P as the initial cluster centers. ; II. Perform Min-Max normalization on each feature dimension of the user preference analysis data P, so that all feature values ​​are scaled to the [0,1] interval: in Let m be the minimum value of the m-th feature among all samples. Let m be the maximum value of the m-th feature in all samples. For user j, the original measurement value on feature m. These are the normalized user feature values. Let m be the original center value of cluster i on feature m. These are the normalized cluster center eigenvalues; III. Calculate the Euclidean distance metric: For each user sample point Calculate the distance from each normalized cluster center: in The value is the Euclidean distance. Let j be the feature vector of the j-th user. Let be the feature vector of the i-th cluster center, n be the dimension of the feature space, and m be the feature dimension index (m = 1, 2, ..., n). Let m be the normalized feature value of user j in the m-th dimension. Let be the normalized center value of the m-th dimension of cluster i; IV. Assigning samples to nearest neighbor clusters: Assigning sample points to nearest neighbor clusters. Assign to the category of the cluster center with the smallest distance; V. Update cluster center locations: Recalculate the center point of each cluster: in For the i-th cluster set, Let i be the new position of the i-th cluster center after the (t+1)th iteration, where i is the cluster center number (i = 1, 2, ..., k). This serves as an identifier for the number of iterations. VI. Iterate through the steps until the cluster center positions converge, and the offset between adjacent iteration centers is less than the threshold. The final user group classification data is output, including age group classification labels, interest clustering identifiers, and high-frequency behavioral feature codes.

[0026] S3 includes the following steps: S31. Obtain user preference analysis data P and audio-visual content data; S32. Match the user preference analysis data P with the pre-stored audio-visual content library, and generate playback control command data based on the matching results. ; When a match is successful, an adaptive playback command is output; If a match fails, the default playback command is output.

[0027] S4 includes the following steps: S41. When the environmental parameter data indicates a high-energy-consumption mode, a dynamic bitrate adjustment algorithm is used to adjust the playback control command data. The resolution parameters are adjusted downward according to the energy-saving threshold, and energy-saving strategy data E is generated. S42. Based on the energy-saving strategy data E, perform power consumption prediction processing to generate the expected energy saving rate data, which represents the percentage of energy savings per unit time.

[0028] S5 includes the following steps: S51. Generate the energy-saving strategy data E and playback control command data. The corresponding control parameters are arranged aligned according to the playback timeline, and the audio-visual playback output data is constructed. : in Output data for audio and video playback. For energy-saving strategy data, For playback control command data; S52, Based on audio / video playback output data The driver performs audio-visual rendering processing on the playback device to generate real-time playback stream data.

[0029] S6 includes the following steps: S61. Establish a standard adjustment strategy matrix corresponding to different user feedback types: in Adjust the strategy matrix to the standard. This represents the adjustment strategy corresponding to the k-th feedback type, including pause strategy, fast forward strategy, and volume adjustment strategy. ; S62, Combine playback adjustment command data with In Perform strategy matching to generate optimal adjustment instruction data.

[0030] S7 includes the following steps: S71, Output data for audio / video playback The playback adjustment command data and system status report data are combined to construct a summary of audio and video playback data. : in Data is compiled for video and audio playback. For system status identification; S72. The intelligent analysis platform calls the monitoring program to perform real-time performance analysis on R based on the system status report data, and generates system health report data.

[0031] Example 1: Adaptive Playback Control Based on Real-Time Emotion Perception In a home theater scenario, users initiate movie playback via voice command. The system uses a built-in microphone to collect ambient noise in real time, while simultaneously capturing facial expression data from the user's camera. The intelligent sensing module's emotion recognition algorithm detects an increase in the frequency of frowning, and combined with physiological data transmitted from a heart rate wristband, determines that the user is in an anxious state. At this point, the user preference analysis module dynamically retrieves historical behavior data from its database, generating a preference tag including "high-stimulation action movies." The playback control module immediately stops the originally scheduled art film, switches to the "Fast & Furious" series, automatically increases the picture contrast, and reduces ambient noise. The entire decision-making response time is less than 40ms, achieving dynamic matching between environmental noise data and the user's immediate emotional state, improving the relevance of the playback content.

[0032] Example 2: Optimization of Data Synchronization for Cross-Device Collaborative Playback Users watch a live football match on the in-vehicle system until halftime. After getting out of the car, they request to continue the playback via a mobile app. The collaborative control module first detects fluctuations in the mobile network bandwidth, synchronously queries the in-vehicle playback data packets stored on the cloud server, dynamically compresses the video stream to 720p based on device performance differences, and separates the audio track into stereo. Simultaneously, it preloads key data frames through distributed caching. When the user clicks play, the system automatically aligns with the in-vehicle system's audio-visual timeline, maintaining a stable 45fps frame rate even under bandwidth fluctuations, with no audio-visual tearing or buffering interruptions throughout, improving the success rate of cross-device switching.

[0033] Example 3: Dynamic Adjustment and Control of Energy Efficiency and Image Quality Balance In outdoor use scenarios with mobile projectors, when the ambient light sensor detects a sudden increase in sunlight intensity, the dynamic balancing module immediately initiates quantitative control: based on hardware temperature data, it maps the optimal energy efficiency solution, first reducing HDR rendering precision to standard mode, then gradually increasing screen brightness while simultaneously disabling background audio surround sound processing. When battery capacity is low, a deep power-saving strategy is triggered, maintaining 4K resolution for key facial areas while reducing other areas to 720p. During this time, the system temperature remains stable below 48℃, extending continuous playback time and improving energy efficiency compared to the default mode.

[0034] Example 4: Dynamic Priority Arbitration and Collaborative Output in Multi-User Scenarios In a video conferencing scenario in a conference room, the system simultaneously detected conflicting behavior from three users: the supervisor initiated a 4K design drawing sharing request via the touch panel, the engineer used voice commands to request playback of the previous frame's technical parameters, and the visitor used gestures to lower the volume. The user analysis module immediately activated a priority arbitration algorithm, automatically allocating a response sequence based on preset role permissions: first, the supervisor's request was locked as the highest priority, and the playback control module activated a local caching acceleration mechanism, instantly outputting the ultra-high-definition design drawing for projection; simultaneously, the engineer's playback command was delayed by 500ms to generate a secondary response, using audio separation technology to separately enhance the volume of the technical parameter channel; finally, the visitor's volume control was executed, intelligently limiting the lower limit of volume reduction to ensure the clarity of the conference audio. The entire process, without upgrading device performance, achieved a multi-task parallel response latency of only 68ms, improving the decision-making efficiency of the conference system.

[0035] Example 5: Disaster Recovery Playback and Cross-Platform Restoration in Extreme Environments When the home theater experiences a power outage, the system's embedded environmental sensing unit instantly detects the voltage anomaly. The emergency handling module immediately triggers a three-level disaster recovery protocol: first, it cuts off power to non-core components—the ambient lighting and Bluetooth speaker group—ensuring the continued operation of the main display device; then, it seamlessly switches to backup power, downgrading the 4K movie to 1080P HDR mode according to a preset strategy; simultaneously, the cloud synchronization module compresses the current playback progress, audio track settings, and user preferences into encrypted data packets and uploads them to the cloud server in real time via a low-power mesh network. Twelve hours later, when the user logs into their account on the subway mobile app, the collaborative control module automatically identifies the phone screen size and network environment, accurately reconstructing the playback state before the power outage from the cloud: precise frame-by-frame image position, ambient volume of 72dB, and the activation status of the director's narration audio track. The entire system minimizes the probability of data loss throughout the entire process from abnormal interruption to cross-platform recovery.

[0036] Example 6: Intelligent Audio-Visual Guidance and Decision Support System for Industrial Environments In the scenario of a large hydropower station equipment monitoring center, operators need to simultaneously monitor multiple equipment operation videos, and the working environment is subject to equipment noise and electromagnetic interference. The system captures abnormal voiceprint characteristics through a distributed microphone array, and the user analysis module identifies specific operator gestures in real time—for example, detecting continuous pressing of the temple indicates a state of inattention. The playback control module immediately initiates a tiered response: dynamically compressing the resolution of non-critical monitoring images while significantly enhancing the display effect of areas with fluctuating parameters; the energy-saving optimization module automatically switches rendering modes based on the main control panel's hardware temperature, using lightweight graphics technology to redraw key data curves to reduce energy consumption; the environmental perception unit captures electromagnetic interference and instantly triggers a synchronization compensation protocol to ensure strict synchronization of audio and video output. When abnormal temperature rise occurs in the equipment, the emergency handling module simultaneously generates multimodal alarm signals: providing voice prompts about the anomaly type and overlaying dynamic warning boxes on the screen. Continuous system operation verification demonstrates that the system intercepts human error, improves the efficiency of critical alarm response, accurately predicts equipment failures, and achieves energy-saving benefits.

[0037] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0038] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An integrated intelligent audio-visual playback system, characterized in that: The system includes: The content acquisition module acquires multi-source audio and video content data from local storage devices, network streaming media services, and cloud databases. Its built-in decoder is compatible with high-definition video streams, 4K video streams, and compressed audio streams, and supports real-time data capture and conversion. The user analysis module, based on deep learning models and natural language processing algorithms, analyzes voice commands, gesture recognition, and touch operation data input from external devices, and generates preference analysis data containing user age groups, interest tags, and viewing frequency parameters through pattern recognition. The playback control module dynamically adjusts the playback order, volume level, and picture quality settings based on this preference analysis data to achieve adaptive playback control. The energy-saving optimization module collects light intensity, ambient temperature and system load parameters through sensors, and uses dynamic bitrate adjustment technology to optimize the power consumption of the processor and display device, reducing screen brightness in low light environment and reducing processor frequency under high temperature conditions. The status monitoring module monitors CPU utilization, memory usage, and network bandwidth parameters in real time. It triggers three levels of alerts through anomaly detection algorithms: Level 1 alerts optimize resource allocation, Level 2 alerts initiate data cleanup, and Level 3 alerts generate a forced hibernation command.

2. The integrated intelligent audio-visual playback system according to claim 1, characterized in that: The system also includes a cloud synchronization module and an emergency response module: The cloud synchronization module establishes an encrypted transmission channel to synchronize playback progress and user preference data to the cloud server in real time, supporting cross-device retrieval and resume playback; The emergency handling module has an embedded fault prediction algorithm. When the system performance reports an abnormality, it automatically switches to the backup playback mode and maintains basic image quality output until the fault is resolved, ensuring playback continuity.

3. The integrated intelligent audio-visual playback system according to claim 1, characterized in that: The system includes the following steps: S1. Obtain audio-visual content data and user input data; S2. Based on the audio-visual content data and the user input data, perform user preference analysis processing to generate user preference analysis data; S3. Based on the user preference analysis data and the audio-visual content data, perform intelligent playback control processing to generate playback control instruction data; S4. Based on the playback control command data and environmental parameter data, perform energy-saving optimization processing to generate energy-saving strategy data; S5. Perform audio-visual playback execution processing based on the energy-saving strategy data and the playback control instruction data to generate audio-visual playback output data; S6. Based on the audio-visual playback output data and user feedback data, perform real-time adjustment processing to generate playback adjustment instruction data; S7. Construct summary data for audio and video playback, perform system status monitoring and processing, and generate system status report data.

4. The integrated intelligent audio-visual playback system according to claim 3, characterized in that: S1 includes the following steps: S11. Extract uncompressed audio and video raw data from the local storage device through the content acquisition module, and generate video data V and audio data A. The video data V represents the combination parameters of frame sequence parameters and resolution parameters established with the time axis as a one-dimensional coordinate axis. The video data includes high-definition video streams and 4K video streams. S12. Capture user interaction data from external devices through the user input module and generate user input data U. The user input data U includes voice command data, gesture recognition data and touch operation data. The external devices include microphone, camera and touch screen.

5. An integrated intelligent audio-visual playback system according to claim 3, characterized in that: S2 includes the following steps: S21. Import the generated video data V, audio data A, and user input data U into the intelligent analysis platform, and use a deep learning algorithm to search for target user behavior feature information in the intelligent analysis platform according to preference keywords, and generate user preference analysis data P, where the unit of P is preference weight value; S22. Based on the user preference analysis data P, the K-means clustering algorithm is used to classify user groups. Euclidean distance is used as the metric to calculate the sample similarity in the feature space, generating user group classification data; wherein: I. Define the initialization rule for cluster centers: Randomly select k sample points from the preference analysis data P as the initial cluster centers. ; II. Perform Min-Max normalization on each feature dimension of the user preference analysis data P, so that all feature values ​​are scaled to the [0,1] interval: in Let m be the minimum value of the m-th feature among all samples. Let m be the maximum value of the m-th feature in all samples. For user j, the original measurement value on feature m. These are the normalized user feature values. Let m be the original center value of cluster i on feature m. These are the normalized cluster center eigenvalues; III. Calculate the Euclidean distance metric: For each user sample point Calculate the distance from each normalized cluster center: in The value is the Euclidean distance. Let j be the feature vector of the j-th user. Let be the feature vector of the i-th cluster center, n be the dimension of the feature space, and m be the feature dimension index (m = 1, 2, ..., n). Let m be the normalized feature value of user j in the m-th dimension. Let be the normalized center value of the m-th dimension of cluster i; IV. Assigning samples to nearest neighbor clusters: Assigning sample points to nearest neighbor clusters. Assign to the category of the cluster center with the smallest distance; V. Update cluster center locations: Recalculate the center point of each cluster: in For the i-th cluster set, Let i be the new position of the i-th cluster center after the (t+1)th iteration, where i is the cluster center number (i = 1, 2, ..., k). This serves as an identifier for the number of iterations. VI. Iterate through the steps until the cluster center positions converge, and the offset between adjacent iteration centers is less than the threshold. The final user group classification data is output, including age group classification labels, interest clustering identifiers, and high-frequency behavioral feature codes.

6. The integrated intelligent audio-visual playback system according to claim 3, characterized in that: S3 includes the following steps: S31. Obtain the user preference analysis data P and the audio-visual content data; S32. Match the user preference analysis data P with the pre-stored audio-visual content library, and generate playback control command data based on the matching results. ; When a match is successful, an adaptive playback command is output; If a match fails, the default playback command is output.

7. The integrated intelligent audio-visual playback system according to claim 3, characterized in that: S4 includes the following steps: S41. When the environmental parameter data indicates a high-energy-consumption mode, a dynamic bitrate adjustment algorithm is used to adjust the playback control command data. The resolution parameters are adjusted downward according to the energy-saving threshold, and energy-saving strategy data E is generated. S42. Based on the energy-saving strategy data E, perform power consumption prediction processing to generate expected energy-saving rate data, which represents the percentage of energy savings per unit time.

8. The integrated intelligent audio-visual playback system according to claim 3, characterized in that: S5 includes the following steps: S51. The generated energy-saving strategy data E and the playback control command data are... The corresponding control parameters are arranged aligned according to the playback timeline, and the audio-visual playback output data is constructed. : in Output data for audio and video playback. For energy-saving strategy data, For playback control command data; S52, Based on the audio / video playback output data The driver performs audio-visual rendering processing on the playback device to generate real-time playback stream data.

9. An integrated intelligent audio-visual playback system according to claim 3, characterized in that: S6 includes the following steps: S61. Establish a standard adjustment strategy matrix corresponding to different user feedback types: in Adjust the strategy matrix to the standard. The adjustment strategies corresponding to the k-th feedback type include pause strategy, fast forward strategy, and volume adjustment strategy. ; S62, Combine the playback adjustment command data with the... In Perform strategy matching to generate optimal adjustment instruction data.

10. An integrated intelligent audio-visual playback system according to claim 3, characterized in that: S7 includes the following steps: S71, output the audio / video playback data The playback adjustment command data and the system status report data are combined to construct a summary of audio and video playback data. : in Data is compiled for video and audio playback. For system status identification; S72. The intelligent analysis platform calls the monitoring program to perform real-time performance analysis on R based on the system status report data, and generates system health report data.

Images