Sound quality improving method, device and equipment of Bluetooth sound box and storage medium

Abstract

The invention relates to the technical field of sound quality enhancement, in particular to a sound quality improvement method and device of a Bluetooth loudspeaker box, equipment and a storage medium. The method comprises the following steps: collecting sound box environment acoustic signals, and carrying out peripheral audio time-frequency spectrum analysis and three-dimensional sound field fitting to construct a peripheral three-dimensional sound field model; based on the surrounding three-dimensional sound field model, environment noise source semantic deep analysis is carried out, and user real-time scene sound quality demand prediction is carried out, so that real-time scene sound quality demand features are generated; bluetooth sound box loudspeaker monitoring parameters are collected, and adaptive noise frequency suppression adjustment is carried out based on the surrounding three-dimensional sound field model rate, so that a loudspeaker dynamic noise suppression adjustment strategy is generated; and acquiring a Bluetooth audio output signal, and performing band frequency response characteristic optimization according to the real-time scene tone quality demand characteristics to obtain a frequency response tone quality optimized audio signal. According to changes of different environments and requirements, the output tone quality of the Bluetooth loudspeaker box is optimized in real time, and high-fidelity audio output is kept.

Description

Technical Field

[0001] The present invention relates to the technical field of sound quality enhancement, and particularly to a method, device, equipment and storage medium for improving the sound quality of a Bluetooth speaker. Background Art

[0002] With the continuous progress of technology and the increasing demand for sound quality by consumers, Bluetooth speakers, as convenient wireless audio playback devices, have been widely used in multiple scenarios such as home entertainment, mobile office, and outdoor activities. Bluetooth speakers, with their characteristics of wireless transmission and easy portability, have become an indispensable part of people's daily lives. However, although Bluetooth speakers have advantages in terms of convenience, due to the influence of various factors such as the size of the space, environmental noise, and the quality of audio signal transmission, their sound quality performance often fails to reach the ideal level. Especially in complex environments, the sound quality of Bluetooth speakers will be affected by factors such as the direction of the sound source, reflection, diffraction, and noise interference, resulting in problems such as audio distortion, low-frequency weakening, or high-frequency distortion, which affect the user's auditory experience.

[0003] Currently, traditional methods for optimizing the sound quality of Bluetooth speakers mainly rely on hardware upgrades, manual adjustments, or fixed audio optimization algorithms, which are usually optimized for specific environments or speaker settings and cannot make flexible adjustments according to the changes in the real-time environment. In addition, traditional sound quality optimization methods cannot take into account the personalized sound quality needs of users, and there are often problems such as inaccurate adjustment and limited sound quality improvement effect. Summary of the Invention

[0004] To solve the above technical problems, the present invention proposes a method, device, equipment and storage medium for improving the sound quality of a Bluetooth speaker to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a method for improving the sound quality of a Bluetooth speaker, including the following steps: Step S1: Collect the acoustic signals of the speaker environment, perform time-frequency spectrum analysis and three-dimensional sound field fitting of the surrounding audio, and construct a surrounding three-dimensional sound field model; Step S2: Based on the surrounding three-dimensional sound field model, perform semantic depth analysis of the environmental noise source and predict the real-time scene sound quality requirements of the user, so as to generate real-time scene sound quality requirement characteristics; Step S3: Collect the monitoring parameters of the Bluetooth speaker's speaker, and perform adaptive noise frequency suppression adjustment based on the rate of the surrounding three-dimensional sound field model, so as to generate a dynamic noise suppression adjustment strategy for the speaker; Step S4: Obtain the Bluetooth audio output signal, and optimize the band frequency response characteristics according to the real-time scene sound quality requirement characteristics to obtain a frequency response sound quality optimized audio signal; Step S5: Mine the spatial acoustics effect of the surrounding three-dimensional sound field model and predict the dynamic spatial audio change to generate the prediction trend of the three-dimensional sound field spatial audio; Step S6: Based on the prediction trend of the three-dimensional sound field spatial audio and the speaker dynamic noise suppression adjustment strategy, perform intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal to execute the sound quality optimization operation of the Bluetooth speaker.

[0006] In this specification, a sound quality improvement device for a Bluetooth speaker is provided, which is used to execute the sound quality improvement method of the Bluetooth speaker as described above, including: A three-dimensional sound field module, which is used to collect the acoustic signals of the speaker environment, perform time-frequency spectrum analysis of the surrounding audio and three-dimensional sound field fitting, and construct a surrounding three-dimensional sound field model; A sound quality requirement module, which is used to perform semantic depth analysis of the environmental noise source based on the surrounding three-dimensional sound field model and predict the real-time scene sound quality requirements of the user, so as to generate the real-time scene sound quality requirement characteristics; A noise suppression module, which is used to collect the monitoring parameters of the Bluetooth speaker and perform adaptive noise frequency suppression adjustment based on the surrounding three-dimensional sound field model rate, so as to generate a speaker dynamic noise suppression adjustment strategy; A sound quality optimization module, which is used to obtain the Bluetooth audio output signal and optimize the band frequency response characteristics according to the real-time scene sound quality requirement characteristics to obtain a frequency response sound quality optimized audio signal; A spatial audio prediction module, which is used to mine the spatial acoustics effect of the surrounding three-dimensional sound field model and predict the dynamic spatial audio change to generate the prediction trend of the three-dimensional sound field spatial audio; An acoustic phase optimization module, which is used to perform intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal based on the prediction trend of the three-dimensional sound field spatial audio and the speaker dynamic noise suppression adjustment strategy to execute the sound quality optimization operation of the Bluetooth speaker.

[0007] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the sound quality improvement method of the Bluetooth speaker described in any one of the above are realized.

[0008] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the sound quality improvement method of the Bluetooth speaker described in any one of the above are realized.

[0009] The beneficial effects of the present invention are specifically as follows: By collecting environmental acoustic signals through a multi-channel microphone array, all surrounding audio signals can be efficiently captured, including background noise, music, conversations, etc. Using time-frequency analysis (such as STFT or Wavelet Transform), more detailed time-domain and frequency-domain features can be extracted from the original signal, providing a reliable basis for subsequent analysis. By fitting the three-dimensional sound field, the direction and distance of noise sources in the environment can be accurately simulated. This can not only help identify various sound sources in the environment but also improve the accuracy of audio processing, making it more refined in subsequent noise suppression and sound optimization. Constructing a three-dimensional sound field model of the surrounding area can accurately locate the azimuth of the sound source, support the implementation of the directional noise suppression algorithm, thereby effectively reducing the impact of environmental noise on the sound quality of the speaker and enhancing the clarity and quality of the audio output. Based on the three-dimensional sound field model, the semantic depth analysis of environmental noise sources can help the system identify different types of noise and classify them (such as human voices, mechanical noise, natural environmental noise, etc.). Through this analysis, the speaker can dynamically adapt to different noise sources and make reasonable adjustments. By analyzing the real-time scene sound quality requirements of users (for example, in a quiet environment, users may need higher low-frequency response, while in a noisy environment, stronger noise suppression is required), the speaker can automatically adjust the sound quality according to different environments and user preferences. This personalized sound quality adjustment enhances the audio experience and is more adaptable. The generation of real-time scene sound quality requirement characteristics helps to achieve dynamic adjustment of audio playback, enabling the speaker to provide the best sound quality in various scenarios and enhancing user satisfaction. By collecting real-time monitoring parameters of the speaker (such as power, damping, etc.), the system can detect the health status of the speaker and ensure its operation in the best working condition. Real-time monitoring helps to identify potential hardware problems in advance and issue warnings. Combining with the three-dimensional sound field model analysis, according to the directivity and frequency characteristics of different noise sources, the noise suppression strategy of the speaker can be adjusted in real time. This not only improves the suppression effect of environmental noise but also avoids excessive damage to the sound quality and enhances the clarity of the audio signal. According to the frequency of surrounding noise sources and the feedback information of the speaker, the working state of the speaker is intelligently adjusted. Different noise frequencies may require different processing methods, and the dynamic noise suppression adjustment strategy of the speaker can enable the speaker to maintain good sound quality performance in various environments. By optimizing the band frequency response characteristics according to the real-time scene sound quality requirements of users, the sound quality performance of different frequency bands can be adjusted. For example, in an indoor environment, strengthening the low-frequency response may be more needed, while in a noisy environment, reducing high-frequency noise is required. By optimizing the frequency response, the speaker can improve the performance of different audio signals. By adjusting the frequency response characteristics according to real-time requirements, it can ensure that the sound quality always matches the user's expectations and environmental conditions. It avoids the inflexibility of static audio adjustment, enabling users to enjoy high-quality audio output in different scenarios.By exploring the spatial acoustic effects of the surrounding three-dimensional sound field model, the system can accurately capture the sound propagation path, reflection, and diffraction characteristics in the environment. This helps to achieve more precise spatial audio processing, such as dynamic sound field adjustment, beamforming, and other technologies, thereby optimizing the user's auditory experience. Dynamic spatial audio change prediction can grasp the trend of sound field changes in the environment in real time (for example, furniture movement, wall reflection, etc.), providing a basis for subsequent sound quality optimization. By predicting the changes in the sound field, the speaker can automatically adjust its audio output when the environment changes, improving the sound effect stability. By integrating the three-dimensional sound field spatial audio prediction trend and the speaker dynamic noise suppression strategy, the system can comprehensively optimize the sound quality at each instant. Whether it is the dynamic adjustment of spatial sound effects or the real-time suppression of noise sources, it can ensure that the speaker provides the best sound quality performance. Based on the changes in different environments and requirements, the speaker can adjust the sound quality optimization strategy in real time to ensure high-fidelity audio output in any situation. Users can enjoy a consistent high-quality sound effect experience in a changing environment. The dynamic adjustment of spatial audio combined with noise suppression and sound quality optimization can bring a more immersive audio experience to users, enhancing their entertainment and work experiences. Description of the Drawings

[0010] Figure 1 It is a schematic flowchart of the steps of a method for improving the sound quality of a Bluetooth speaker according to the present invention; Figure 2 It is a schematic flowchart of the detailed implementation steps of step S1; Figure 3 It is a schematic flowchart of the detailed implementation steps of step S2; Figure 4 It is a schematic flowchart of the detailed implementation steps of step S3. Detailed Implementation Manner

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0012] The embodiments of the present application provide a method, device, equipment, and storage medium for improving the sound quality of a Bluetooth speaker. The execution subjects of the method, device, equipment, and storage medium include but are not limited to: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc. that carry this system, which can be regarded as general computing nodes of the present application. The data processing platform includes but is not limited to at least one of: an audio image management system, an information management system, and a cloud data management system.

[0013] Please refer to Figures 1 to 4 , the present invention provides a method for improving the sound quality of a Bluetooth speaker, and the method for improving the sound quality of the Bluetooth speaker includes the following steps: Step S1: Collect the ambient acoustic signals of the speaker, perform time-frequency spectrum analysis and three-dimensional sound field fitting on the surrounding audio, and construct a surrounding three-dimensional sound field model; Step S2: Based on the surrounding three-dimensional sound field model, perform in-depth semantic analysis of the environmental noise source and predict the real-time scene sound quality requirements of the user, so as to generate real-time scene sound quality requirement features; Step S3: Collect the monitoring parameters of the Bluetooth speaker's speaker, and perform adaptive noise frequency suppression adjustment based on the surrounding three-dimensional sound field model rate, so as to generate a dynamic noise suppression adjustment strategy for the speaker; Step S4: Obtain the Bluetooth audio output signal, and optimize the band frequency response characteristics according to the real-time scene sound quality requirement features to obtain a frequency response sound quality optimized audio signal; Step S5: Mine the spatial acoustics effect of the surrounding three-dimensional sound field model and predict the dynamic spatial audio change to generate a three-dimensional sound field spatial audio prediction trend; Step S6: Based on the three-dimensional sound field spatial audio prediction trend and the speaker dynamic noise suppression adjustment strategy, perform intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal to execute the sound quality optimization operation of the Bluetooth speaker.

[0014] In the embodiment of the present invention, refer to Figure 1 , which is a schematic diagram of the step flow of a method for improving the sound quality of a Bluetooth speaker according to the present invention. In this example, the steps of the method for improving the sound quality of the Bluetooth speaker include: Step S1: Collect the ambient acoustic signals of the speaker, perform time-frequency spectrum analysis and three-dimensional sound field fitting on the surrounding audio, and construct a surrounding three-dimensional sound field model; In this embodiment, ensure that the required equipment, such as a professional microphone array, an audio interface, and data acquisition software (such as Audacity or MATLAB), is properly prepared. Select a microphone with high sensitivity to capture subtle sounds in the environment. Usually, a microphone with a frequency response range of 20 Hz to 20 kHz is selected. Configure the acquisition system and set the sampling rate (usually 48 kHz or 96 kHz) and bit depth (16 bits or 24 bits) to ensure high-quality signal acquisition. Arrange the microphones around the speakers to form a reasonable array (such as circular or rectangular) to capture acoustic information from different directions. Conduct audio signal acquisition under different environmental conditions (such as indoor, outdoor, silent, and background noise environments). It is recommended that each recording duration be 5 to 10 minutes to ensure that various sound sources and background noises are covered. Record the time, environmental conditions, and background noise level of each acquisition for subsequent analysis. During the recording process, ensure that the volume is kept within an appropriate range (such as -10 dB to 0 dB) to avoid signal distortion. Use a signal monitoring tool to observe the audio signal in real time to ensure data quality. Apply the short-time Fourier transform (STFT) or wavelet transform (WT) to convert the time-domain signal into a time-frequency domain signal. Selecting the wavelet transform can better capture the instantaneous characteristics of non-stationary signals, while STFT is suitable for the analysis of uniformly frequency signals. Set the analysis parameters, including the window function type (such as Hanning window, Blackman window, etc.), window length (usually 256 to 1024 points), and overlap rate (usually 50% overlap) to ensure the best time-frequency resolution. Conduct time-frequency spectrum analysis on the acquired environmental acoustic signals to generate a time-frequency diagram. Through STFT or wavelet transform, analyze the frequency components and intensity distribution at each moment. Record the energy and frequency changes in each frequency band, especially focusing on the frequency bands sensitive to the human ear (such as 300 Hz to 3 kHz). Generate a visualization chart to display the time-frequency diagram for easy analysis and comparison of the spectral characteristics at different time periods. Heat maps or three-dimensional graphs can be used to show the relationship between frequency and time to ensure clear and readable information. Organize the analyzed time-frequency spectrum characteristics into a data table to ensure clear traceability of information. Each data point should include time, frequency, and energy values for subsequent sound field model construction. Generate a report recording the results and conclusions of the time-frequency spectrum analysis to provide a basis for subsequent three-dimensional sound field fitting. Select a sound field modeling technique based on ray tracing (RayTracing) or the finite element method (FiniteElementMethod). These methods can simulate the propagation, reflection, and attenuation processes of sound waves in the environment to create an accurate three-dimensional sound field model. Set the model parameters, including the sound source position, receiver position, wall material, room size, and geometry, to ensure the accuracy and repeatability of the model. Use the collected time-frequency spectrum data to fit the sound field model. Calculate the propagation of sound waves along different paths through the ray tracing algorithm, analyze the phenomena of reflection, diffraction, and interference, and generate a sound field distribution map.Record the sound pressure level changes, frequency responses, and other acoustic characteristics in the sound field model to ensure that the model can accurately reflect the acoustic characteristics in the actual environment. Visualize the generated three-dimensional sound field model to ensure the intuitiveness of the information. The characteristics such as the sound field distribution and sound pressure level changes can be displayed through 3D graphics software for in-depth analysis and research. Generate a detailed report of the sound field model, recording the parameter settings, fitting results during the model construction process, and their significance in practical applications, providing support for subsequent acoustic research and applications.

[0015] Step S2: Based on the surrounding three-dimensional sound field model, perform semantic deep analysis of the environmental noise sources and predict the user's real-time scene sound quality requirements, thereby generating real-time scene sound quality requirement characteristics; In this embodiment, a suitable noise source identification technology is selected, such as machine learning algorithms (e.g., support vector machine (SVM), random forest, or convolutional neural network (CNN)), to achieve in-depth analysis of different environmental noise sources. These technologies can identify noise sources through feature extraction and classification. Key parameters are set, such as the size of the training set, feature extraction methods (e.g., MFCC, Chroma features), and noise source categories (e.g., traffic noise, conversation sound, music sound, etc.), to ensure the accuracy and reliability of the identification. Environmental noise signals are extracted from the previously constructed three-dimensional sound field model, and the sound pressure levels and time characteristics in different frequency bands are recorded. The collected signals are preprocessed, including denoising, framing, and window function processing, to prepare the data for feature extraction. The short-time Fourier transform (STFT) or wavelet transform is used to perform time-frequency analysis on the noise signals to extract the time-frequency characteristics of each noise source for classification and identification. The selected machine learning model is used to train and test the extracted features to identify the main noise sources in the environment. In the training phase, a labeled data set is used to improve the classification ability of the model. The recognition accuracy and confusion matrix of each noise source are recorded to evaluate the recognition effect of different noise sources. The model parameters are adjusted to optimize the recognition performance to ensure that different types of noise sources can be accurately distinguished. A suitable real-time sound quality demand prediction model is selected, such as the ARIMA model based on time series analysis or the recurrent neural network (RNN) based on deep learning. These models can predict the sound quality demands of users in specific scenarios by analyzing historical data and real-time data. Model parameters are set, such as input features, historical data length, and number of training epochs, to ensure the effectiveness and accuracy of the model. Real-time audio playback data of users are collected through the built-in sensors of Bluetooth headsets or speakers, including the type of playback content, volume settings, environmental noise levels, etc. The sampling frequency is set (e.g., collect data once per second) to capture changes in user behavior. The collected data is preprocessed, including normalization and denoising, to improve the data quality and the prediction ability of the model. The timestamp and environmental conditions of each collection are recorded for subsequent analysis. The selected prediction model is trained using historical data to identify the sound quality demand characteristics of users in different scenarios. Cross-validation is performed to ensure the generalization ability and accuracy of the model. In the real-time scenario, the newly collected data is input into the trained model to generate a prediction of the user's sound quality demand. The prediction results are recorded, including the required sound quality characteristics (e.g., frequency response, volume, equalization settings, etc.). The real-time scenario sound quality demand characteristics generated are organized into a data table to ensure the structuring of information. The table should include time, predicted sound quality characteristics, and their corresponding environmental conditions for subsequent analysis and optimization. Visualization charts are generated to show the changing trends of the user's sound quality demands, intuitively presenting the sound quality demand characteristics in different scenarios, and providing a basis for subsequent sound quality adjustment and optimization.

[0016] Step S3: Collect the monitoring parameters of the Bluetooth speaker's speaker, and perform adaptive noise frequency suppression adjustment based on the surrounding three-dimensional sound field model rate, so as to generate a dynamic noise suppression adjustment strategy for the speaker; In this embodiment, ensure that the speaker of the Bluetooth speaker and its monitoring devices (such as sensors and microphones) are working properly. Select high-sensitivity sensors (such as accelerometers and sound pressure sensors) to facilitate the acquisition of the operating parameters of the speaker. Configure the data acquisition system, set the sampling frequency (usually select 48 kHz or 96 kHz) and bit depth (such as 24 bits) to ensure that high-frequency and low-frequency sounds emitted by the speaker can be captured. Set a reasonable measurement range according to the experimental environment (for example, set the maximum sound pressure level to 90 dB). When the speaker plays different audio signals (such as white noise, music, speech, etc.), perform real-time acquisition of monitoring parameters. The recorded parameters include input voltage, output current, displacement of the speaker, frequency response, and ambient noise level, etc. Each signal is collected for at least 5 minutes to ensure data diversity. Record the timestamp, environmental conditions, and audio signal type for each acquisition for subsequent analysis. Ensure that the acquisition is carried out under different background noise conditions to evaluate the performance of the speaker in various environments. Use the previously constructed three-dimensional sound field model to analyze the distribution of noise sources and the characteristics of noise propagation in the environment. Determine the main noise frequency components and their impact on the speaker output. Set the parameters of the sound field model, including the sound source position, receiving point position, wall material, and room geometry, to ensure that the model truly reflects the environmental characteristics. Select a suitable adaptive noise suppression algorithm, such as an adaptive filter (such as the LMS or RLS algorithm), to process the collected speaker signals. These algorithms can dynamically adjust the filter parameters according to real-time environmental feedback. Set the key parameters of the algorithm, such as the step size, filter order, and target frequency range, to ensure the effectiveness of noise suppression. While real-time monitoring the speaker output, apply the adaptive noise suppression algorithm to adjust the output signal of the speaker in real time. By analyzing the phase and amplitude differences between the environmental noise and the speaker output, adjust the output signal to reduce the noise impact. Record the parameters and effects of each adjustment, such as the noise suppression ratio and the user's sound quality feedback, to evaluate the effectiveness and real-time response ability of the algorithm. Organize the results of adaptive noise frequency suppression into a data table to ensure the clarity and traceability of the information. The table should include the noise suppression effects under different conditions, the adjusted parameters, and their corresponding environmental conditions for subsequent analysis and optimization. Generate a report, record the implementation process, effect evaluation of the speaker dynamic noise suppression adjustment strategy, and its significance in practical applications, to provide support for subsequent sound quality optimization. According to the collected monitoring data and the results of adaptive noise suppression, formulate a dynamic noise suppression adjustment strategy for the speaker. Ensure that the strategy can automatically adjust according to environmental changes to provide the best sound quality experience. Set the trigger conditions for dynamic adjustment, such as environmental noise level, speaker output characteristics, and user feedback, to achieve intelligent noise suppression. In practical applications, implement the formulated dynamic noise suppression adjustment strategy, and real-time monitor the speaker performance and user feedback to ensure the effectiveness and reliability of the strategy.Establish a feedback mechanism to feedback the user's sound quality feedback and environmental change information into the adjustment strategy to achieve continuous optimization and improvement.

[0017] Step S4: Obtain the Bluetooth audio output signal, and optimize the band frequency response characteristics according to the real-time scene sound quality requirement characteristics to obtain the frequency response sound quality optimized audio signal; In this embodiment, ensure that the Bluetooth speaker or headset is properly connected to the sound source device (such as a mobile phone or computer), and make settings to ensure the best sound quality. Select a high-quality sound source device to avoid signal loss during audio transmission. Configure the software of the sound source device to ensure that the audio output format is of high quality (for example, select aptX or AAC encoding) to ensure the integrity of the audio signal. Use professional audio acquisition software (such as Audacity or MATLAB), set the sampling rate (usually 48 kHz or 96 kHz) and bit depth (16 bits or 24 bits) to ensure that high-quality audio signals can be captured. Adopt a high-sensitivity microphone or audio interface to capture the Bluetooth audio output signal in real time. During the acquisition process, ensure that the volume remains within a reasonable range (such as -10 dB to 0 dB) to avoid distortion. In different playback scenarios (such as music, dialogue, ambient sounds, etc.), perform real-time acquisition of audio signals. Record the audio data in each scenario, and it is recommended that the acquisition duration be 5 to 10 minutes to ensure that diverse audio characteristics are covered. Record the time, environmental conditions, and playback content of each acquisition for subsequent analysis. Ensure that the acquired signal is not interfered by external noise, and adopt noise suppression measures if necessary to improve the acquisition quality. Based on the real-time scene sound quality requirement characteristics extracted in the foregoing steps, analyze the user's sound quality preferences in different scenarios. Machine learning algorithms (such as K-means clustering) can be used to classify the user preferences and extract key sound quality characteristics (such as clarity, low-frequency response, and high-frequency brightness). Record the sound quality requirement characteristics in each scenario, including the required frequency range, gain, and phase response, etc., for subsequent optimization use. Select a suitable frequency response optimization algorithm, such as a digital equalizer (EQ) or an adaptive filter, to optimize the frequency response characteristics of the audio signal. The digital equalizer can optimize the frequency response characteristics of the audio signal by adjusting the gain of different frequency bands. Set the frequency band parameters of the equalizer (such as low frequency, middle frequency, high frequency bands), usually divide the frequency band into 10 or more frequency bands to ensure the fineness of optimization. Input the obtained Bluetooth audio output signal into the frequency response optimization algorithm and process it according to the real-time scene sound quality requirement characteristics. Adjust the gain of each frequency band to make the output signal meet the user's sound quality requirements. During the optimization process, monitor the spectral changes of the audio signal in real time to ensure that the optimized signal achieves the best effect within the target frequency range. Record the gain value and effect of each adjustment for subsequent analysis. Compare the optimized frequency response sound quality signal with the original signal to analyze the optimization effect. Use objective indicators (such as root mean square error, frequency response curve) and subjective tests (such as listening evaluation) to verify the effectiveness of the optimization. Generate an optimization report, record the process, effect, and user feedback of the frequency response optimization, and provide a basis for subsequent sound quality improvement.

[0018] Step S5: Mine the spatial acoustics effect and predict the dynamic spatial audio change of the surrounding three-dimensional sound field model to generate the three-dimensional sound field spatial audio prediction trend; In this embodiment, the mined acoustic effect data is combined with the real-time audio change data to prepare a training dataset for dynamic spatial audio change prediction. Ensure that the dataset contains various environmental conditions and different audio signals to improve the generalization ability of the model. Select a suitable prediction model, such as a long short-term memory network (LSTM) or a time series prediction model (such as ARIMA), to capture the changing trend of audio signals over time. Use the prepared dataset to train the selected prediction model. Set training parameters, including the learning rate, batch size, and number of training epochs, to ensure that the model can effectively learn the temporal features in the data. Conduct cross-validation to evaluate the accuracy and stability of the model, ensuring that the model can perform effective audio change prediction in different scenarios. Record the loss value and accuracy rate during the training process to adjust the model parameters. Input the real-time audio and environmental change data into the trained model to predict the changing trend of spatial audio in real time. Record the prediction results for each time period, including audio features (such as frequency response, loudness change, etc.). Monitor the accuracy of the prediction results and adjust the input features of the model in real time to improve the prediction accuracy. A threshold can be set to trigger retraining of the model when the prediction error exceeds a certain range.

[0019] Step S6: Based on the three-dimensional sound field spatial audio prediction trend and the speaker dynamic noise suppression adjustment strategy, perform intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal to execute the sound quality optimization operation of the Bluetooth speaker.

[0020] In this embodiment, the frequency response optimized audio signal after previous optimization is integrated with the three-dimensional sound field spatial audio prediction trend data. Ensure that the data formats of the two are consistent for subsequent processing. Generally, the frequency response optimized audio signal should have a high sampling rate (such as 96 kHz) and a high bit depth (such as 24 bits) to ensure the integrity of the sound quality. Using audio processing software (such as MATLAB or Audacity), align the frequency response optimized audio signal with the prediction trend data. Determine the time window length and overlap rate to retain the necessary time information during dynamic optimization. According to the previously formulated speaker dynamic noise suppression adjustment strategy, analyze the current ambient noise level and speaker output characteristics. The dynamic noise suppression strategy should adjust parameters based on real-time monitoring data, such as the noise frequency range and suppression gain. Collect real-time ambient noise data and prepare the parameters for dynamic adjustment. These data include the current noise spectrum, sound pressure level (SPL), and user feedback information to ensure that environmental changes are reflected in real time during the optimization process. Select appropriate dynamic sound quality optimization algorithms, such as adaptive equalizer (EQ) and dynamic range compression (DRC) techniques. The adaptive equalizer can dynamically adjust the frequency response according to real-time audio input and environmental feedback, while dynamic range compression can control the dynamic range of the audio signal to improve the overall sound quality experience. Set the key parameters of the equalizer, including frequency band division (such as low frequency, mid frequency, high frequency), gain adjustment range (usually set to ±6 dB), and the threshold (such as -10 dB) and ratio (such as 4:1) of dynamic range compression to ensure the flexibility and adaptability of sound quality optimization. During audio playback, monitor the frequency response optimized audio signal in real time and apply the dynamic optimization algorithm. Through real-time analysis of the audio signal, adjust the gain settings of the equalizer to achieve the best frequency response characteristics. For example, for low-frequency signals, it may be necessary to increase the gain to enhance the bass effect, while for high-frequency signals, the gain needs to be moderately reduced to avoid harshness. Record the parameters and effects of each adjustment, such as the change in frequency band gain, the impact of dynamic compression, and the user's sound quality feedback. Ensure that the processing can respond to the user's needs and environmental changes in real time. During the implementation of dynamic optimization, regularly evaluate the effect of sound quality optimization. Use objective evaluation indicators (such as total harmonic distortion (THD), signal-to-noise ratio (SNR)) and subjective listening tests (such as user evaluation) for comprehensive evaluation. According to the evaluation results, further adjust the parameters of the dynamic optimization algorithm. For example, if the user feedback indicates that the low frequency is too strong, it may be necessary to moderately reduce the low-frequency gain or increase the high-frequency gain to achieve a more balanced sound quality effect.

[0021] In this embodiment, refer to Figure 2 , which is a schematic diagram of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Collect the acoustic signal of the speaker environment based on the multi-channel microphone array built in the Bluetooth speaker; Perform a time-frequency spectrum analysis on the ambient acoustic signal of the speaker to generate the time-frequency spectrum characteristics of the ambient audio; Identify ambient noise based on the ambient acoustic signal of the speaker and mark multiple ambient noise points; Calculate the sound source direction and distance of the multiple ambient noise points; Perform an analysis of the spatial distribution of the noise points based on the sound source direction and distance to generate a spatial distribution map of the ambient noise points; Perform a three-dimensional sound field fitting on the spatial distribution map of the ambient noise points based on the time-frequency spectrum characteristics of the ambient audio to construct a surrounding three-dimensional sound field model.

[0022] In this embodiment, ensure that the multi-channel microphone array built into the Bluetooth speaker works properly. Calibrate the microphone to ensure that its sensitivity and frequency response meet the experimental requirements. Generally, the selected frequency range should cover the audible audio frequency range of the human ear (20 Hz - 20 kHz) to obtain comprehensive environmental acoustic signals. Configure the recording parameters of the Bluetooth speaker, including the sampling frequency (usually select 48 kHz or 96 kHz), bit depth (16 bits or 24 bits), and recording duration. Ensure that the settings meet the requirements of audio signal processing. Record audio signals under different environmental conditions (such as indoor, outdoor, noisy environments, etc.). Ensure that a long enough recording period (recommended 5 - 10 minutes) is recorded in each environment to capture diverse acoustic characteristics. Record the environmental conditions of each recording, including temperature, humidity, and background noise level, etc., for reference during subsequent analysis. Store the collected audio signals in a high-capacity storage device to ensure that the data is not lost. Use a lossless format (such as WAV or FLAC) for storage to retain the sound quality. Perform preliminary data processing to remove silent segments and obvious interference signals to ensure the data quality for subsequent analysis. Select a suitable time-frequency analysis method, such as the Short-Time Fourier Transform (STFT) or Wavelet Transform (WT), to convert the time-domain signal into a time-frequency domain signal. These methods can effectively capture the instantaneous characteristics and frequency changes of the signal. Set the analysis parameters, such as the window function type (Hanning window, Blackman window, etc.), window length, and overlap rate (usually select 50% overlap) to ensure the best time-frequency resolution. Perform time-frequency spectrum analysis on the collected environmental acoustic signals to generate a time-frequency diagram. Use professional audio analysis software (such as MATLAB, SciPy library in Python, etc.) for processing. Extract important features from the time-frequency diagram, such as band energy, frequency distribution, and time characteristics, etc., and record the values and change trends of each feature. Visualize the generated time-frequency spectrum diagram for intuitive analysis. Display the relationship between frequency and time through a heat map or three-dimensional graph to ensure that the information is clear and easy to read. Record the results of the time-frequency spectrum analysis, including the energy distribution, peak frequency, and corresponding time period of each frequency band, to ensure that the algorithm can effectively distinguish environmental noise from background sound. Set the threshold parameters for noise identification, usually based on the previous time-frequency spectrum analysis results. The optimal threshold can be determined through experiments to reduce the situation of misidentification. Run the noise identification algorithm to analyze the environmental acoustic signals, automatically detect and mark multiple environmental noise points. Record the start time, end time, and corresponding frequency characteristics of each noise point. Conduct manual review to ensure the accuracy of the markings and make fine-tuning if necessary to eliminate mislabeled noise points. Organize the detected environmental noise points into a data table to ensure the information is structured. The table should include information such as the timestamp, frequency, duration, and intensity of the noise points. Adopt a sound source localization algorithm, such as the Beamforming technology of the multi-channel microphone array or the Time Difference of Arrival (TDOA) method, to calculate the sound source direction and distance of each noise point.Configure the geometric structure parameters of the microphone array, such as microphone spacing and array shape (linear, circular, etc.), to ensure positioning accuracy. For each marked environmental noise point, use the selected sound source localization algorithm to calculate its sound source direction (expressed in degrees) and distance (in meters). Record the localization results of each noise source to ensure the accuracy of the data. Conduct multiple experiments to verify the localization effect under different conditions (such as environmental changes, microphone placement angles) to optimize the algorithm performance. Organize the calculated sound source direction and distance information into a data table to ensure the information is clear and understandable. The table should include information such as the ID, direction angle, distance, and intensity of each noise point. Select a suitable spatial analysis method, such as Kriging interpolation or heat map generation technology, to analyze the spatial distribution characteristics of the noise points. Ensure that the method can effectively display the concentration degree and distribution trend of the noise points. Determine the analysis parameters, such as interpolation range, grid size, etc., to ensure the resolution and readability of the spatial distribution map. Based on the calculated sound source direction and distance information, conduct a spatial distribution analysis of the noise points. Use the selected spatial analysis method to generate a spatial distribution map of the noise points. Record the position of each noise point on the spatial distribution map and its corresponding intensity for subsequent analysis and comparison. Visualize the generated spatial distribution map of environmental noise points to ensure clear information display. The noise intensity can be represented by color gradients or icon sizes for easy observation of the noise distribution. Write an analysis report to record the results and conclusions of the spatial distribution analysis, providing a reference for subsequent noise control and environmental optimization. Select a suitable three-dimensional sound field modeling technology, such as the finite element method (FEM) or the boundary element method (BEM), to convert the spatial distribution information of environmental noise points into a three-dimensional sound field model. Configure the model parameters, such as the size of the calculation grid and acoustic characteristics, to ensure the accuracy and calculation efficiency of the model. Based on the time-frequency spectrum characteristics of the surrounding audio and the spatial distribution information of environmental noise points, conduct a fitting calculation of the three-dimensional sound field. Use the selected modeling method to generate a three-dimensional sound field model. Record the parameter settings and results of each calculation step for subsequent analysis and optimization. Verify the generated three-dimensional sound field model to ensure that the model can accurately reflect the sound field characteristics in the actual environment. The reliability of the model can be evaluated by comparing it with actual measurement data. Visualize the three-dimensional sound field model to ensure the intuitiveness of the information. Use three-dimensional graphics software to display characteristics such as sound field distribution and sound pressure level changes for in-depth analysis and research.

[0023] The specific embodiments are as follows: Experimental Environment and Equipment Configuration Speaker Equipment: The Bluetooth speaker is equipped with four microphone arrays (the microphones are separated by 90°), and the effective frequency response range of each microphone is from 20 Hz to 20 kHz.

[0024] Microphone Array Configuration: The microphone array is arranged in a ring with a diameter of approximately 10 cm. The sampling frequency of each microphone is 44.1 kHz to ensure high-fidelity audio acquisition.

[0025] Ambient Noise Simulation: Multiple noise sources are set up in the laboratory, including traffic noise, air conditioner sounds, and human voices. The average volume of the noise in the environment is 50 dB, and the maximum noise can reach 80 dB.

[0026] Temperature and Humidity: The experiment is conducted in a normal temperature (about 22°C) environment, and the humidity is maintained between 40% and 60%.

[0027] Parameter Settings: Number of Microphone Arrays: 4 Sampling Frequency: 44.1 kHz (meeting the high-precision requirements of audio signals) Signal Duration: Each signal acquisition lasts for 10 seconds, and the experiment is repeated 5 times to increase reliability.

[0028] Data Recording: The data collected each time is transmitted to the computer via Bluetooth, and the data format is WAV (lossless format).

[0029] Analysis Methods and Parameters: Method: Short-Time Fourier Transform (STFT) is used to perform time-frequency analysis on the collected signals.

[0030] Window Function: Hamming window, with a window length of 2048 points and a frame overlap rate of 50% (i.e., 1024 points) to balance time and frequency resolution.

[0031] Frequency Range: The frequency range of the analyzed signal is from 20 Hz to 20 kHz.

[0032] Spectrum Resolution: The resolution of each frequency point is 21.5 Hz (through 2048-point STFT conversion).

[0033] The time-frequency spectrum diagram is obtained through STFT, and the noise points with higher energy in the spectrum are marked. These points may represent surrounding noise sources.

[0034] Analyze the energy peaks in the time-frequency spectrum, set an energy threshold (such as 5 dB higher than the background noise), and identify noise sources through this threshold.

[0035] Clustering Algorithm: DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is used to classify noise points. By setting a distance threshold (such as 5 ms) and the minimum number of noise points (such as 3), different noise sources are identified.

[0036] Energy Threshold: 5 dB (compared to background noise) Clustering Parameters: Distance threshold 5 ms, minimum number of points 3 Clustering algorithm: DBSCAN According to the clustering results, the experiment marks multiple noise source points, corresponding to noises in different directions and frequency bands respectively.

[0037] Calculate the direction and distance of the noise source through the time difference of receiving the same noise source by the microphone array. Assuming the geometric position of the microphone array is known, calculate the sound source direction through triangulation.

[0038] Calculation formula: Use the time difference between microphones and the known speed of sound (343 m / s) to calculate the distance from the sound source to the microphone array.

[0039] Microphone array spacing: 10 cm Speed of sound: 343 m / s (at room temperature) Time difference accuracy: Assuming the time difference accuracy is 0.1 ms, the sound source direction accuracy can reach 1°.

[0040] Calculate the relative direction and distance of each noise source. For example, a noise source may be located 30° in front of the speaker, 2 meters away from the speaker; another noise source may be located 60° to the right of the speaker, 1.5 meters away from the speaker. By visualizing the direction and distance data of each noise source, generate a spatial distribution map of noise points. Analyze the density and distribution law of noise points to reveal the concentrated area and diffusion area of noise. Use Matplotlib or other visualization tools to generate a three-dimensional noise point distribution map, where each point in the figure represents the position of a noise source, and the color represents the noise intensity. Set the spatial resolution to 0.5 meters to accurately depict the distribution of noise sources. Noise intensity attenuation model: Calculate the noise intensity according to the distance attenuation formula, assuming the initial intensity of each noise source is 75 dB. Generate a three-dimensional space map to show the spatial distribution of noise sources. For example, it is found that most noise sources are concentrated within the range of 2 to 3 meters directly in front of the speaker, and the noise is weak in the surrounding area. Based on the spatial distribution and time-frequency characteristics of noise sources, use the finite element method (FEM) or boundary element method (BEM) to model the sound field in the environment. Modeling process: By simulating the reflection, refraction, diffraction and other phenomena of sound wave propagation, fit out the sound field model in the environment. The model includes the influence of different noise sources on each position in space.

[0041] Finite element method (FEM) Calculation accuracy: The accuracy requirement of the sound field model is 1 cm, and the calculation grid size is 0.5 cm.

[0042] Reflection and absorption coefficient: The wall absorption coefficient set in the experiment is 0.3, and the ground absorption coefficient is 0.5.

[0043] In this embodiment, refer to Figure 3, which is a schematic diagram of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Performing semantic depth analysis on environmental noise sources based on the surrounding three-dimensional sound field model to generate environmental noise semantic features; Identifying the current scene of the environmental noise semantic features to obtain the playing scene of the current Bluetooth speaker; Obtaining the historical speaker monitoring log, and mining the user's personalized habits for the playing scene of the current Bluetooth speaker according to the historical speaker monitoring log to generate the user's personalized scene habits; Predicting the user's real-time scene sound quality requirements based on the user's personalized scene habits, so as to generate real-time scene sound quality requirement features.

[0044] In this embodiment, a three-dimensional sound field model is used to first deeply analyze environmental noise sources. Through the data in the sound field model, the acoustic characteristics of different noise sources are analyzed, including information such as frequency, volume, and phase. Parsing parameters are set, such as the frequency range (20 Hz - 20 kHz) and the sound pressure level (SPL) threshold, to ensure coverage of the audible range of the human ear and important sound source characteristics. A detailed spectral analysis is performed on each sound source to extract its characteristic parameters. Based on the noise characteristics generated by the sound field model, machine learning algorithms (such as support vector machines, convolutional neural networks, etc.) are used for semantic feature extraction. These algorithms can identify specific sound patterns from acoustic signals, such as human voices, traffic noise, music, etc. The semantic labels and their characteristic values of each noise source are recorded to construct an environmental noise semantic feature library. The time-frequency spectral characteristics of the audio can be combined to ensure that the extracted features can accurately reflect the semantic information of the noise source. The extracted environmental noise semantic features are visually displayed to generate corresponding feature maps for subsequent analysis. Heat maps or three-dimensional graphs are used to display the feature distributions of different noise sources to ensure that the information is intuitive and easy to understand. A suitable scene recognition algorithm is selected, such as a deep learning-based image classification algorithm or audio classification algorithm, to ensure effective differentiation of different playback scenes (such as indoor, outdoor, party, etc.). The training data set required for scene recognition is collected and prepared, including audio samples in various environments, to ensure the comprehensiveness and accuracy of training. The collected audio data is used for model training, and the algorithm parameters are optimized to improve the recognition accuracy. The cross-validation method is used to evaluate the model performance to ensure effective recognition in different scenes. Training parameters are set, such as the learning rate, batch size, and number of training epochs, to ensure that the training process of the model can converge and achieve the expected recognition effect. The extracted environmental noise semantic features are input into the trained model for real-time scene recognition. According to the output of the model, the current playback scene of the Bluetooth speaker is determined. The recognition results are recorded, including the recognized scene type and its confidence level, for subsequent user habit mining and sound quality requirement prediction. Historical speaker monitoring logs are extracted from the Bluetooth speaker, including the user's playback records, volume settings, playback time, and environmental information, etc. These data will provide a basis for user habit analysis. Parameters such as the timestamp, audio source, and environmental noise level of each log entry are recorded to ensure the integrity and accuracy of the data. Data mining techniques, such as clustering analysis and association rule mining, are used to analyze the historical monitoring logs to identify the user's personalized scene habits. For example, the K-means clustering method is used to classify the playback scenes to find the scene types preferred by the user. Mining parameters are set, such as the number of clusters and the similarity threshold, to ensure the effectiveness and accuracy of habit analysis. User personalized scene habit characteristics are generated according to the analysis results, and information such as the user's sound quality preference, volume selection, and playback time in different environments is recorded. The results of the user habit analysis are sorted out and a report is generated for subsequent sound quality requirement prediction and personalized service provision.In the current playback scenario, the user's audio playback behavior is monitored in real time, and data such as volume, playback time, and audio type are collected. This data will be used for predicting the user's sound quality requirements. Record the environmental characteristics (such as noise level, space type, etc.) of each playback to ensure the accuracy and relevance of the real-time data. Based on the user's personalized scenario habits, establish a sound quality requirement prediction model. This can be modeled using regression analysis, time series prediction, or deep learning models (such as LSTM) to predict the user's sound quality requirements in the current scenario. Set model parameters, such as input features, learning rate, and number of training epochs, to ensure that the model can accurately reflect the user's sound quality preferences. Input the real-time collected data into the prediction model to generate the sound quality requirement characteristics of the user in the current scenario. These characteristics can include the ideal volume, equalization settings, and sound quality modes, etc. Record the prediction results to ensure the accuracy of the requirement characteristics and generate a real-time report for subsequent sound quality optimization and personalized settings.

[0045] In this embodiment, refer to Figure 4 , which is a schematic diagram of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Collect the monitoring parameters of the Bluetooth speaker; calculate the driving power and damping of the monitoring parameters of the Bluetooth speaker; Conduct time-series vibration feature analysis based on the driving power and damping to generate the vibration characteristics of the speaker's operating state; Calculate the frequency of each noise point of the surrounding three-dimensional sound field model to generate the noise point frequencies in different directions; Conduct adaptive noise frequency suppression calculation on the noise point frequencies in different directions to generate the multi-directional noise frequency suppression values; Adjust the dynamic vibration parameters of the speaker's operating state vibration characteristics according to the multi-directional noise frequency suppression values, thereby generating a dynamic noise suppression adjustment strategy for the speaker.

[0046] In this embodiment, ensure that the speaker of the Bluetooth speaker and its monitoring devices (such as sensors, microphones) work properly. Select an appropriate type of sensor (such as an accelerometer, microphone array) to facilitate the acquisition of the operating parameters of the speaker. Configure the parameters of the acquisition system, including the sampling frequency (recommended 48 kHz or higher), to ensure that high-frequency and low-frequency sounds emitted by the speaker can be captured. Set an appropriate measurement range to adapt to the output power of the speaker. Under the condition that the speaker plays different audio signals (such as white noise, music, conversation, etc.), perform real-time acquisition of monitoring parameters. The recorded parameters include input voltage, output current, displacement of the speaker, frequency response, and temperature, etc. To ensure the accuracy of the data, perform multiple acquisitions, record the timestamp, environmental conditions, and audio signal type for each acquisition for subsequent analysis and comparison. According to the acquired voltage and current data, calculate the driving power of the speaker. Use the formula P = V×I (power = voltage × current) to calculate the instantaneous power at each time point. Perform time-domain analysis of the power, record the average power and peak power under different audio signal conditions to ensure the accuracy of the power calculation. For example, experimental parameters can be set to calculate the power range when playing different audio signals. According to the displacement response and input power of the speaker, calculate the damping coefficient of the speaker. By analyzing the difference between the frequency response of the speaker and the actual output, apply the damping formula D = Calculate using / v (damping = force / velocity). Record the damping values at each frequency and compare them with the standard values to ensure the accuracy and reliability of the calculations. Organize the calculated driving power and damping values into a data table to ensure the clarity and traceability of the information. The table should include key parameters such as power, damping value, and frequency for subsequent vibration characteristic analysis. Generate a report summarizing the power and damping characteristics under different audio signal conditions to provide a reference basis for subsequent analysis. Select a suitable vibration characteristic analysis method, such as time-domain analysis, frequency-domain analysis, or wavelet transform, to facilitate the extraction of the vibration characteristics of the speaker. Ensure that the selected method can effectively capture the dynamic response of the speaker. Set the analysis parameters, such as window size and overlap rate, to ensure the best time-frequency resolution for the analysis. Preprocess the collected vibration data, including denoising, smoothing, etc., to ensure the data quality. Use a filter (such as a low-pass filter) to remove high-frequency noise for better analysis of the vibration characteristics. Perform vibration characteristic extraction and record key parameters such as vibration amplitude, frequency components, and phase information to generate the operating-state vibration characteristics of the speaker. Visualize the analyzed vibration characteristics to generate corresponding characteristic diagrams, such as vibration spectrograms or time-domain waveforms, for easy analysis. Use heat maps or three-dimensional graphs to display the vibration characteristics at different frequencies and times to ensure that the information is intuitive and easy to read. Organize the results of the vibration characteristic analysis and form a report recording the operating-state vibration characteristics of the speaker and their changing trends to provide a basis for subsequent noise suppression calculations. Use the data in the sound field model to calculate the frequency for each noise point. Select a suitable calculation method, such as Fourier transform or time-frequency analysis, to extract the frequency information of the noise points. Set the calculation parameters, such as sampling frequency and analysis window, to ensure the accuracy of the frequency calculation. In the three-dimensional sound field model, calculate the frequency for each noise point one by one and record the frequency characteristics of each point. Use the sound pressure level data of the sound field model and combine with the acoustic characteristics of the environment to determine the frequency range of the noise source. Record the frequency values of each noise point and compare them with other noise points to analyze the noise characteristics in different directions. Organize the calculated frequency values of the noise points in different directions into a data table to ensure the structuring of the information. The table should include information such as the ID, direction, and frequency of the noise points for subsequent analysis. Visualize the noise point frequency data to generate a frequency distribution diagram or a heat map to intuitively display the noise characteristics in different directions for further noise suppression calculations. Select a suitable adaptive noise suppression algorithm, such as an adaptive filter (LMS or RLS algorithm), to dynamically adjust the noise frequency suppression value. These algorithms can respond to environmental changes in real time and automatically optimize the suppression effect. Set the algorithm parameters, including step size, filter order, etc., to ensure that the algorithm can work effectively under different conditions. Perform adaptive noise suppression calculations on the calculated noise point frequencies to generate the noise frequency suppression values in each direction. Adjust the filter parameters according to the noise frequency characteristics to achieve the best suppression effect.Record the suppression values in each direction, compare them with the original noise frequencies, and evaluate the improvement in suppression effect. Organize the generated multi-directional noise frequency suppression values into a data table to ensure clear information. The table should include information such as the ID of the noise point, direction, suppression value, etc., for subsequent analysis. Compile an analysis report to record the results and conclusions of the noise frequency suppression calculation, providing a basis for subsequent dynamic vibration parameter adjustment. Adopt a dynamic vibration parameter adjustment method based on feedback control. According to the generated noise frequency suppression values, adjust the driving parameters of the speaker (such as input voltage, audio signal processing) in real time. Set the parameters of the adjustment strategy, including the adjustment amplitude and adjustment frequency, to ensure the response speed and accuracy. Dynamically adjust the operating vibration characteristics of the speaker according to the multi-directional noise frequency suppression values. Monitor the vibration feedback of the speaker in real time, adjust the driving power and damping to optimize the vibration characteristics. Record the parameters and effects of each adjustment to ensure the traceability and repeatability of the dynamic adjustment process. Organize the generated dynamic noise suppression adjustment strategy into a document to ensure the structuring and clarity of the information. The document should include the background of the adjustment strategy, implementation steps, and effect evaluation for subsequent application and optimization. Conduct an effect evaluation, record the adjusted vibration characteristics and noise suppression effect to verify the effectiveness of the strategy and provide a reference for future improvement.

[0047] The specific embodiments are as follows: Select a Bluetooth speaker equipped with a speaker with an impedance of 4Ω, a rated power of 10W, and a maximum power of 20W. The speaker is built-in with an acceleration sensor (MEMS accelerometer) and current and voltage sensors to monitor the vibration and electrical parameters of the speaker.

[0048] Acquisition device: Use a high-precision data acquisition card (such as NI USB-6341) with a sampling accuracy of 16 bits and a maximum sampling frequency of 100kHz for real-time data acquisition. The sampling frequency of the acceleration sensor is 10kHz, and the sampling frequencies of current and voltage are 1kHz.

[0049] Noise source: Set noise sources in the indoor environment, such as air conditioners, televisions, traffic noise, etc., with a noise intensity range of 40dB to 80dB. The distribution of the noise sources is monitored through multiple microphone arrays.

[0050] The main parameters collected include: speaker input voltage (0V to 15V), input current (0A to 5A), vibration acceleration (0g to 50g).

[0051] The sampling accuracy is: Voltage: accuracy 0.1V, sampling frequency 1kHz Current: accuracy 0.01A, sampling frequency 1kHz Vibration: accuracy 0.1g, sampling frequency 10kHz The Hall sensor is used to collect voltage and current data in real time, and the data is transmitted to a computer through a data acquisition card for processing.

[0052] The accelerometer is installed at the center of the speaker diaphragm to collect vibration signals. The collection time is 10 seconds, and the experiment is carried out 5 rounds to improve the reliability of the data.

[0053] Calculation method of driving power: P = V×I, where P is power, V is voltage, and I is current. Assuming the voltage of the speaker is 12V and the current is 1.2A, the calculated power is: P = 12V×1.2A = 14.4W. After calculating the driving power, the power consumption of the speaker is obtained.

[0054] The simplified Lambert damping model is adopted, and the damping force is calculated through the vibration signal (acceleration) and vibration velocity: F(d)=−c×v, where F(d) is the damping force, c is the damping coefficient, and v is the vibration velocity. Assuming the damping coefficient c is 0.05N·s / m and the vibration velocity is 5cm / s, the calculated damping force is: F(d)=−0.05×0.05 = −0.0025N. The driving power is 14.4W and the damping force is -0.0025N. This data helps to analyze the energy efficiency and vibration attenuation characteristics of the speaker.

[0055] The collected vibration signal is analyzed in the frequency domain through FFT to extract the vibration spectrum of the speaker. The sampling frequency is 10kHz, the FFT length is 1024 points, and the frequency resolution is 9.77Hz (10kHz / 1024).

[0056] In the time-domain analysis, first, the vibration mode of the speaker in the working state is analyzed by calculating the characteristics such as the average amplitude and peak value of the vibration signal.

[0057] In the frequency-domain analysis, by calculating the vibration intensity in different frequency bands, the main vibration frequency of the speaker and its harmonic components are obtained. In the time-domain analysis, the peak amplitude of the vibration signal is 5g, the average amplitude is 2g, and the main frequency components in the spectrum are 300Hz, 600Hz, and 1200Hz (these are the fundamental frequency of the speaker and its second and third harmonics).

[0058] The high-frequency noise (such as above 2kHz) accounts for a relatively small proportion in the vibration signal, and the low-frequency (such as 100Hz - 500Hz) vibration intensity is relatively large, indicating that the main vibration energy of the speaker is concentrated in the low-frequency range.

[0059] In the three-dimensional sound field model, the frequency distribution of the noise source is closely related to the working frequency of the speaker and its harmonics. Assuming the working frequency of the speaker is 1kHz, the frequency distribution of the surrounding noise points is analyzed.

[0060] By calculating the distance, direction, frequency of each noise point and its relationship with the operating frequency of the speaker, the frequency characteristics of each noise point are obtained.

[0061] The noise points are distributed in the range from 1 meter to 3 meters away from the speaker. A circular array (such as a 360-degree array) is used to monitor the noise frequencies in different directions. Assuming the frequency range of each noise point is from 100 Hz to 10 kHz, the spectrum of each point is analyzed using FFT. Multiple noise point frequencies are calculated, including the frequencies of 3 noise points at 500 Hz, 1 kHz, and 2 kHz being 800 Hz, 1.1 kHz, and 1.9 kHz respectively. The LMS (Least Mean Square) adaptive filtering algorithm is used to adjust the filter parameters to maximize the suppression of the noise frequencies. The filter is adaptively adjusted according to the different noise frequencies and directions.

[0062] Assuming the noise frequencies are 500 Hz, 1 kHz, and 2 kHz, calculate their corresponding noise frequency suppression values.

[0063] The goal of the adopted noise suppression algorithm is to reduce the noise to the lowest point. For the noise at 1 kHz frequency, the noise frequency suppression value is -20 dB.

[0064] For the noises at 500 Hz and 2 kHz, the suppression values are -15 dB and -18 dB respectively. According to the adaptive noise frequency suppression results, the vibration parameters of the speaker are adjusted. The adjustment methods include adjusting the driving power and damping coefficient of the speaker to reduce unnecessary vibrations.

[0065] Assume the adjustment range of the driving power of the speaker is from 5 W to 20 W, and the adjustment range of the damping coefficient is from 0.01 N·s / m to 0.1 N·s / m. Adjust the corresponding vibration mode according to the frequency suppression value. The noise suppression effect at 1 kHz frequency is the most significant. After adjustment, the driving power is 12 W, the damping coefficient is adjusted to 0.05 N·s / m, and the vibration amplitude is reduced to 3 g.

[0066] At the noise points of 500 Hz and 2 kHz frequencies, after adjustment, the vibration amplitude and noise intensity are also effectively reduced, and the noise frequency suppression value is further optimized.

[0067] In this embodiment, step S4 includes the following steps: Obtain the Bluetooth audio output signal; perform multi-band decomposition on the Bluetooth audio output signal to generate multiple independent audio frequency band segments; Calculate the band frequency components of the multiple independent audio frequency band segments to generate the audio frequency components of each band; Perform in-depth detection of sound quality degradation based on the audio frequency components of each band, and extract the inferior audio bands; Perform dynamic frequency range compression on the inferior audio band to generate a frequency-compressed inferior audio band; Optimize the band frequency response characteristics of the frequency-compressed inferior audio band according to the real-time scene sound quality requirement characteristics to obtain a frequency response sound quality optimized audio signal.

[0068] In this embodiment, ensure that the Bluetooth speaker or earphone is properly connected to the sound source device (such as a mobile phone or a computer). Set the audio output format of the sound source device to high quality (for example, select aptX or AAC encoding) to ensure the integrity of the audio signal. Select a suitable recording software or audio analysis tool to ensure that it can capture the Bluetooth audio output signal in real time, and set the sampling rate to 44.1 kHz or 48 kHz to meet the CD audio quality standard. When playing different types of audio content (such as music, dialogue, ambient sounds), record the Bluetooth audio output signal in real time. Ensure that the acquisition time period is long enough (recommended 5 - 10 minutes) to capture diverse audio characteristics. Record the environmental conditions, sound source type, and playback volume for each acquisition for reference in subsequent analysis. The volume range can be set (such as -10 dB to 0 dB) to avoid audio distortion. Select a suitable multi - band decomposition technique, such as Wavelet Transform or Filter Bank, to decompose the audio signal into multiple independent frequency bands. Ensure that the selected method can effectively capture the time - frequency characteristics of the audio signal. Set the decomposition parameters, such as the wavelet basis function, decomposition level, or the center frequency of the filter, to ensure the accuracy of the decomposition result. Perform multi - band decomposition on the acquired Bluetooth audio signal to generate multiple independent audio frequency band segments. Record the frequency range and corresponding energy distribution of each band. Conduct multiple decomposition experiments and adjust the decomposition parameters to ensure that sufficient band information can be extracted for subsequent frequency component calculation. Organize the multiple independent audio frequency band segments obtained from the decomposition into a data table to ensure the clarity and traceability of the information. Each band should contain information such as the frequency range, energy distribution, and its relationship with the original signal to generate a report, record the characteristics of different frequency bands and their roles in the audio signal, and provide a basis for subsequent analysis. Select a method for calculating frequency components, such as Fast Fourier Transform (FFT) or Short - Time Fourier Transform (STFT), to extract the frequency components of each band. Ensure that the selected method can effectively analyze the spectral characteristics of the audio signal. Set the calculation parameters, such as the window size and overlap rate (usually select 50% overlap) to ensure the accuracy and resolution of the frequency component calculation. Calculate the frequency components for each independent audio frequency band segment and record the energy values of each band at different frequencies. Generate a spectrogram for each band through FFT or STFT analysis. Record the frequency components of each band, including the main frequency, frequency range, and corresponding energy distribution for subsequent analysis. Organize the calculated audio frequency components of each band into a data table to ensure the structuring of the information. Each frequency band should contain frequency, energy value, and related characteristics for subsequent detection of sound quality degradation.Adopt a sound quality evaluation algorithm, such as PESQ (Perceptual Evaluation of Speech Quality) or POLQA (Perceptual Objective Listening Quality Assessment), to deeply detect the audio frequency components to identify the frequency bands with deteriorated sound quality. Set evaluation parameters, such as evaluation model selection and quality threshold, to ensure the accuracy of deteriorated detection. Conduct sound quality deterioration detection on the audio frequency components of each frequency band, and extract the inferior audio frequency bands. Record the sound quality scores of each frequency band and their corresponding frequency ranges. Conduct multiple detections to ensure accurate identification of inferior frequency bands and avoid false detections. Organize the detected inferior audio frequency bands into a data table to ensure the clarity and traceability of information. The table should contain information such as the ID of the frequency band, frequency range, inferiority score, etc., for subsequent analysis. Generate a report to record the results and conclusions of the sound quality deterioration detection, providing a basis for subsequent frequency range compression. Select a suitable dynamic frequency compression algorithm, such as dynamic range compression (DRC) or adaptive equalizer (EQ), to process the inferior audio frequency bands. Ensure that the algorithm can effectively compress the frequency range and improve the sound quality. Set compression parameters, such as threshold, ratio, and attack / release time, to ensure that the effect of dynamic compression meets the expectations. Conduct dynamic frequency range compression processing on the extracted inferior audio frequency bands to generate inferior audio frequency bands after frequency compression. Monitor the processing effect in real time to ensure that the sound quality is improved after compression. Record the characteristic changes before and after compression of each frequency band to ensure the effectiveness and stability of dynamic compression. Organize the inferior audio frequency bands after frequency compression into a data table to ensure the structuring of information. Each frequency band should contain information such as the frequency range, energy distribution, and sound quality score before and after compression. Analyze the user's preferences and environmental factors (such as noise level, type of playback content) according to the sound quality requirement characteristics of the real-time scenario, providing a basis for subsequent optimization of the frequency response characteristics. Record the user's sound quality preference data, such as commonly used volume settings, frequency response curves, etc., to ensure the pertinence of the optimization plan. Select a suitable frequency response characteristic optimization algorithm, such as equalizer optimization or adaptive filtering algorithm, to adjust the audio frequency bands after frequency compression according to the user's needs. Ensure that the optimization method can effectively improve the sound quality. Set optimization parameters, including the target frequency response curve, gain, and frequency range, to ensure that the optimization effect meets the user's needs. Conduct frequency response characteristic optimization processing on the inferior audio frequency bands after frequency compression to generate the final optimized audio signal with frequency response sound quality. Evaluate the optimization effect in real time to ensure a significant improvement in sound quality. Record the sound quality scores and characteristic changes before and after optimization, and generate a comparison report to verify the effectiveness and feasibility of the optimization strategy.

[0069] In this embodiment, step S5 includes the following steps: Conduct audio reflection evolution analysis on the surrounding three-dimensional sound field model to obtain audio reflection evolution sub-characteristics; Calculate the audio attenuation rate according to the surrounding three-dimensional sound field model and extract the spatial audio attenuation rate; Identify the spatial reverberation time according to the surrounding three-dimensional sound field model to generate the spatial reverberation time; Mine the spatial acoustics effects of the audio reflection evolution sub-features, spatial audio attenuation rate, and spatial reverberation time to generate the three-dimensional sound field spatial acoustics effects; Analyze the change of the environmental sound field based on the surrounding three-dimensional sound field model and extract the change characteristics of the environmental sound field; Predict the dynamic spatial audio change of the three-dimensional sound field spatial acoustics effects according to the change characteristics of the environmental sound field to generate the three-dimensional sound field spatial audio prediction trend.

[0070] In this embodiment, select a suitable analysis method, such as ray tracing (RayTracing) or finite element analysis (Finite Element Analysis), to analyze the reflection and attenuation process of the audio signal in the environment. These methods can help identify the evolution characteristics of the audio under different reflection paths. Set the reflection parameters, such as the number of reflections, the acoustic wave frequency range, and the time step, to ensure the accuracy of the analysis. Conduct an audio reflection evolution analysis on the sound field model, record the intensity, delay, and frequency change of the reflected signal. Generate the reflection evolution characteristics, including key parameters such as reflection time delay, reflection intensity, and frequency response. Organize the analysis results, organize the audio reflection evolution sub-features into a data table to ensure the structuring of information and facilitate subsequent audio attenuation rate calculation and reverberation time identification. Select a suitable attenuation rate calculation method, such as the direct measurement method or the method based on spectral analysis, to ensure that the energy loss of the acoustic wave during propagation can be accurately evaluated. Set relevant parameters, such as the frequency range, measurement distance, and time window, to ensure the accuracy and reliability of the calculation. In the sound field model, calculate the audio attenuation rate at different frequencies. By comparing the sound pressure levels at the sound source and the receiving point, apply the formula: 𝛼 = (Attenuation rate = (Initial sound pressure level - Final sound pressure level) divided by distance), thus calculating the attenuation rate. Record the attenuation values for each frequency and analyze the influence of different environmental conditions (such as wall materials, sound absorption characteristics) on the attenuation rate to ensure the representativeness of the results. Organize the calculated spatial audio attenuation rates into a data table to ensure the clarity and traceability of the information. The table should include the frequency, attenuation rate, and their corresponding environmental conditions for subsequent analysis. Generate a report recording the results and conclusions of the audio attenuation rate calculation to provide a basis for subsequent reverberation time identification. Adopt a suitable reverberation time calculation method, such as the Sabine formula or the Eyring formula, to evaluate the reverberation characteristics of the space. These methods calculate the reverberation time based on the room volume and sound absorption characteristics. Set the experimental parameters, including the room volume and the sound absorption coefficient of the surface material, to ensure the accuracy of the calculation. According to the parameters of the sound field model, use the selected reverberation time calculation formula to calculate the reverberation time of the space. Record the reverberation times at different frequencies to obtain comprehensive reverberation characteristics. Conduct multiple measurements to verify the reverberation times under different conditions to ensure the reliability of the results. Organize the calculated spatial reverberation times into a data table to ensure the structuring of the information. Each frequency band should include the frequency, reverberation time, and their corresponding environmental parameters. Generate a report recording the results and conclusions of the reverberation time identification to provide a basis for subsequent exploration of spatial acoustical effects. Adopt a comprehensive analysis method to combine the characteristics of audio reflection evolution, spatial audio attenuation rate, and reverberation time for the exploration of spatial acoustical effects. Multivariate regression analysis or machine learning models can be used to evaluate the influence of each factor on the acoustical effects. Set the analysis parameters, such as model selection, feature selection, and evaluation metrics, to ensure the scientificity and accuracy of the analysis. Through the analysis of the collected data, identify the influence of audio reflection, attenuation, and reverberation on spatial acoustical effects. Generate acoustical effect features, including key indicators such as clarity, loudness, and spatial sense. Record the values of each feature and conduct a comparative analysis to determine the changes in acoustical effects under different environmental conditions. Organize the analyzed three-dimensional sound field spatial acoustical effects into a data table to ensure the clarity and readability of the information. The table should include the acoustical effect features and their corresponding environmental conditions for subsequent analysis. Generate visualization charts, such as acoustical effect distribution maps, to visually display the acoustical characteristics under different environments and provide a basis for subsequent analysis of sound field changes. Select a suitable sound field change analysis method, such as time-domain analysis or frequency-domain analysis, to evaluate the change characteristics of the environmental sound field. Ensure that the method can effectively capture the dynamic changes of the sound field. Set the analysis parameters, such as the time window, analysis frequency range, and sampling rate, to ensure the accuracy of the results. Conduct an analysis of the environmental sound field change of the sound field model and record the change characteristics of the sound field at different time periods or under different environmental conditions. Generate environmental sound field change features, including key parameters such as sound pressure level change and frequency response change. Conduct multiple analyses to ensure that the influence of environmental changes on the sound field can be captured and record the specific circumstances of the changes.Organize the extracted environmental sound field change features into a data table to ensure the structuring of information. Each feature should include the time, the change amplitude, and its corresponding environmental conditions for facilitating subsequent audio prediction analysis. Generate a report recording the results and conclusions of the environmental sound field change analysis to provide a basis for subsequent spatial audio change prediction. Adopt methods based on time series analysis or machine learning to establish a dynamic spatial audio change prediction model. ARIMA (Autoregressive Integrated Moving Average Model) or LSTM (Long Short-Term Memory Network) can be used for prediction. Set model parameters such as the size of the training dataset, feature selection, and evaluation criteria to ensure the accuracy of prediction. Based on the extracted environmental sound field change features, input them into the prediction model to generate the prediction trend of the three-dimensional sound field spatial audio. Record the prediction results for each time period to analyze the dynamic changes of the sound field. Conduct model validation by comparing with actual measurement data to evaluate the accuracy and reliability of the prediction. Organize the generated prediction trend of the three-dimensional sound field spatial audio into a data table to ensure the clarity and readability of the information. The table should include the time, the predicted values, and their corresponding environmental conditions.

[0071] In this embodiment, step S6 includes the following steps: Based on the three-dimensional sound field spatial audio prediction trend, conduct a dynamic change acoustic phase deviation analysis on the frequency response and sound quality optimized audio signal to generate the acoustic phase deviation under the audio trend; Based on the acoustic phase deviation under the audio trend, conduct phase deviation compensation optimization and perform adaptive beamforming processing to construct a high-fidelity spatial optimization strategy; Based on the high-fidelity spatial optimization strategy and the speaker dynamic noise suppression adjustment strategy, conduct intelligent sound quality dynamic optimization to execute the sound quality optimization operation of the Bluetooth speaker.

[0072] In this embodiment, spatial audio prediction trend data is extracted from the previous three-dimensional sound field model. These data should include features such as frequency response curves, sound pressure level changes, and reverberation time to ensure that they can reflect the real audio change trends. Ensure the integrity and accuracy of the data, record the audio feature changes in different time periods for subsequent phase deviation analysis. Select a suitable phase deviation analysis method, such as Fourier transform (FFT) or phase spectrum analysis, to calculate the acoustic phase deviation at different frequencies. These methods can effectively capture the phase information of the signal in the frequency domain. Set the calculation parameters, especially the sampling frequency (recommended to be 48 kHz or higher) and window size, to ensure the accuracy of phase analysis. Perform dynamic change acoustic phase deviation analysis on the audio signal with optimized frequency response. By comparing the phase information before and after optimization, record the phase difference at each frequency point. Generate a phase deviation map to visually display the phase changes at different frequencies. Record the phase deviation values of each frequency band and mark the significant deviation points for subsequent compensation optimization. Adopt a phase compensation algorithm, such as minimum phase filtering or phase response matching, to compensate for the phase deviation of the audio signal. These algorithms can optimize the quality of the audio signal by adjusting the phase response. Set the compensation parameters, such as compensation amplitude and frequency range, to ensure the effectiveness of the compensation effect. Based on the calculated phase deviation, implement phase compensation optimization processing. By adjusting the phase response of the audio signal, eliminate unnecessary phase deviations and optimize the sound quality. Record the audio feature changes before and after compensation, especially the frequency response and sound quality score, to verify the effectiveness of the compensation effect. Combine the phase compensation results and implement adaptive beamforming technology to optimize the sound field distribution of the speaker. By adjusting the drive signal of the speaker, achieve the enhancement of the target sound source and the suppression of background noise. Set the beamforming parameters, such as beam width and directivity, to ensure that the processing effect meets the actual requirements. Record the sound quality changes under different beamforming strategies. According to the results of compensation and beamforming, construct a high-fidelity spatial optimization strategy. Ensure that the strategy can dynamically respond to environmental changes and provide the best sound quality experience. Adopt intelligent sound quality optimization algorithms, such as deep learning models or adaptive filtering algorithms, combine the high-fidelity spatial optimization strategy with the speaker dynamic noise suppression strategy to achieve dynamic optimization of sound quality. Set the optimization parameters, such as learning rate, number of training epochs, and input features, to ensure the effectiveness and accuracy of the algorithm. Combine the high-fidelity spatial optimization strategy with the speaker dynamic noise suppression strategy for intelligent sound quality dynamic optimization processing. Real-time monitor the audio signal and automatically adjust the sound quality parameters to adapt to different playback scenarios. Record the sound quality feature changes before and after optimization, especially the frequency response, distortion rate, and user feedback, to verify the effectiveness of the optimization effect. Evaluate the optimized sound quality, use objective evaluation indicators (such as THD, SNR) and subjective listening tests, collect user feedback to ensure that the optimization effect meets user expectations. Establish a feedback mechanism to feedback the user's feedback and environmental change information into the optimization algorithm to achieve continuous sound quality improvement.Organize the results of sound quality optimization into a document to ensure the clarity and traceability of information. The report should include content such as the optimization process, effect evaluation, and user feedback for subsequent improvement and optimization. Record the parameter settings and results of each optimization for future reference and adjustment.

[0073] In this embodiment, a sound quality improvement device for a Bluetooth speaker is provided, which is used to execute the sound quality improvement method of the Bluetooth speaker as described above, including: A three-dimensional sound field module, which is used to collect the acoustic signals of the speaker environment, perform time-frequency spectrum analysis of the surrounding audio and three-dimensional sound field fitting, and construct a surrounding three-dimensional sound field model; A sound quality requirement module, which is used to perform in-depth semantic analysis of environmental noise sources based on the surrounding three-dimensional sound field model and predict the real-time scene sound quality requirements of users, so as to generate real-time scene sound quality requirement characteristics; A noise suppression module, which is used to collect the monitoring parameters of the Bluetooth speaker's speaker and perform adaptive noise frequency suppression adjustment based on the surrounding three-dimensional sound field model rate, so as to generate a dynamic noise suppression adjustment strategy for the speaker; A sound quality optimization module, which is used to obtain the Bluetooth audio output signal and optimize the band frequency response characteristics according to the real-time scene sound quality requirement characteristics to obtain a frequency response sound quality optimized audio signal; A spatial audio prediction module, which is used to mine the spatial acoustic effects of the surrounding three-dimensional sound field model and predict the dynamic spatial audio changes to generate a three-dimensional sound field spatial audio prediction trend; An acoustic phase optimization module, which is used to perform intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal based on the three-dimensional sound field spatial audio prediction trend and the speaker dynamic noise suppression adjustment strategy to execute the sound quality optimization operation of the Bluetooth speaker.

[0074] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the sound quality improvement method of the Bluetooth speaker described in any one of the above are implemented.

[0075] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the sound quality improvement method of the Bluetooth speaker described in any one of the above are implemented.

[0076] Those skilled in the art clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, systems, and units described above refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0077] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, is embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (such as a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that store program codes.

[0078] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to cover all changes that fall within the meaning and scope of the equivalent elements of the application documents within the present invention.

[0079] As described above, these are only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A method for improving the sound quality of a Bluetooth speaker, characterized in that, It includes the following steps: Step S1: Collect the acoustic signals of the speaker environment, perform time-frequency spectrum analysis of the surrounding audio and three-dimensional sound field fitting, and construct a surrounding three-dimensional sound field model; Step S2: Based on the surrounding three-dimensional sound field model, perform in-depth semantic analysis of the environmental noise sources, and predict the real-time scene sound quality requirements of the user, so as to generate the real-time scene sound quality requirement features; Step S3: Collect the monitoring parameters of the Bluetooth speaker's speaker, and perform adaptive noise frequency suppression adjustment based on the surrounding three-dimensional sound field model rate, so as to generate a dynamic noise suppression adjustment strategy for the speaker; Step S4: Obtain the Bluetooth audio output signal, and optimize the band frequency response characteristics according to the real-time scene sound quality requirement features to obtain a frequency response sound quality optimized audio signal; Step S5: Mine the spatial acoustic effects of the surrounding three-dimensional sound field model and predict the dynamic spatial audio changes to generate a three-dimensional sound field spatial audio prediction trend; Step S6: Based on the three-dimensional sound field spatial audio prediction trend and the speaker dynamic noise suppression adjustment strategy, perform intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal to execute the sound quality optimization operation of the Bluetooth speaker.

2. The method for improving the sound quality of the Bluetooth speaker according to claim 1, wherein, The specific steps of Step S1 are: Collect the acoustic signals of the speaker environment based on the multi-channel microphone array built in the Bluetooth speaker; Perform time-frequency spectrum analysis of the acoustic signals of the speaker environment to generate the time-frequency spectrum characteristics of the surrounding audio; Identify the environmental noise according to the acoustic signals of the speaker environment, and mark multiple environmental noise points; Calculate the sound source direction and distance of the multiple environmental noise points; Perform spatial distribution analysis of the noise points according to the sound source direction and distance to generate a spatial distribution map of the environmental noise points; Perform three-dimensional sound field fitting on the spatial distribution map of the environmental noise points based on the time-frequency spectrum characteristics of the surrounding audio to construct a surrounding three-dimensional sound field model.

3. The method for improving the sound quality of the Bluetooth speaker according to claim 1, characterized in that, The specific steps of Step S2 are: Perform in-depth semantic analysis of the environmental noise sources based on the surrounding three-dimensional sound field model to generate environmental noise semantic features; Perform current scene recognition on the environmental noise semantic features to obtain the playback scene of the current Bluetooth speaker; Obtain the historical speaker monitoring log, and mine the user's personalized habits for the playback scene of the current Bluetooth speaker according to the historical speaker monitoring log to generate the user's personalized scene habits; Predict the real-time scene sound quality requirements of the user based on the user's personalized scene habits, so as to generate the real-time scene sound quality requirement features.

4. The method for improving the sound quality of the Bluetooth speaker according to claim 1, characterized in that, The specific steps of Step S3 are: Collect the monitoring parameters of the Bluetooth speaker's speaker; calculate the driving power and damping of the monitoring parameters of the Bluetooth speaker's speaker; Perform time-series vibration feature analysis according to the driving power and damping to generate the vibration features of the speaker's operating state; Calculate the noise point frequency for each noise point of the surrounding three-dimensional sound field model to generate the noise point frequencies in different directions; Perform adaptive noise frequency suppression calculation on the noise point frequencies in different directions to generate multi-directional noise frequency suppression values; Adjust the dynamic vibration parameters of the speaker's operating state vibration features according to the multi-directional noise frequency suppression values, so as to generate a dynamic noise suppression adjustment strategy for the speaker.

5. The method for improving the sound quality of the Bluetooth speaker according to claim 1, wherein The specific steps of Step S4 are: Obtain the Bluetooth audio output signal; perform multi-band decomposition on the Bluetooth audio output signal to generate multiple independent audio frequency band segments; Calculate the band frequency components of multiple independent audio frequency bands to generate the audio frequency components of each band; Perform in-depth detection of sound quality degradation based on the audio frequency components of each band, and extract the inferior audio bands; Perform dynamic frequency range compression on the inferior audio bands to generate frequency-compressed inferior audio bands; Optimize the band frequency response characteristics of the frequency-compressed inferior audio bands according to the real-time scene sound quality requirement characteristics to obtain the frequency response sound quality optimized audio signal.

6. The method for improving the sound quality of the Bluetooth speaker according to claim 1, wherein, The specific steps of step S5 are as follows: Perform audio reflection evolution analysis on the surrounding three-dimensional sound field model to obtain the audio reflection evolution characteristics; Calculate the audio attenuation rate according to the surrounding three-dimensional sound field model, and extract the spatial audio attenuation rate; Identify the spatial reverberation time according to the surrounding three-dimensional sound field model to generate the spatial reverberation time; Mine the spatial acoustical effects of the audio reflection evolution characteristics, spatial audio attenuation rate and spatial reverberation time to generate the three-dimensional sound field spatial acoustical effects; Perform environmental sound field change analysis based on the surrounding three-dimensional sound field model, and extract the environmental sound field change characteristics; Predict the dynamic spatial audio change of the three-dimensional sound field spatial acoustical effects according to the environmental sound field change characteristics to generate the three-dimensional sound field spatial audio prediction trend.

7. The method for improving the sound quality of the Bluetooth speaker according to claim 1, wherein The specific steps of step S6 are as follows: Perform dynamic change acoustic phase deviation analysis on the frequency response sound quality optimized audio signal based on the three-dimensional sound field spatial audio prediction trend to generate the acoustic phase deviation under the audio trend; Perform phase deviation compensation optimization based on the acoustic phase deviation under the audio trend, and perform adaptive beamforming processing to construct a high-fidelity spatial optimization strategy; Perform intelligent sound quality dynamic optimization based on the high-fidelity spatial optimization strategy and the speaker dynamic noise suppression adjustment strategy to execute the sound quality optimization operation of the Bluetooth speaker.

8. An audio quality improvement device for a Bluetooth speaker, characterized in that, For executing the sound quality improvement method of the Bluetooth speaker as described in claim 1, including: A three-dimensional sound field module for collecting the acoustic signals of the speaker environment, performing surrounding audio time-frequency spectrum analysis and three-dimensional sound field fitting, and constructing a surrounding three-dimensional sound field model; A sound quality requirement module for performing in-depth semantic analysis of the environmental noise source based on the surrounding three-dimensional sound field model, and predicting the user's real-time scene sound quality requirements, thereby generating real-time scene sound quality requirement characteristics; A noise suppression module for collecting the monitoring parameters of the Bluetooth speaker's speaker, and performing adaptive noise frequency suppression adjustment based on the surrounding three-dimensional sound field model rate, thereby generating a speaker dynamic noise suppression adjustment strategy; A sound quality optimization module for obtaining the Bluetooth audio output signal, and optimizing the band frequency response characteristics according to the real-time scene sound quality requirement characteristics to obtain the frequency response sound quality optimized audio signal; A spatial audio prediction module for mining the spatial acoustical effects and predicting the dynamic spatial audio change of the surrounding three-dimensional sound field model to generate a three-dimensional sound field spatial audio prediction trend; An acoustic phase optimization module for performing intelligent sound quality dynamic optimization on the frequency response sound quality optimized audio signal based on the three-dimensional sound field spatial audio prediction trend and the speaker dynamic noise suppression adjustment strategy to execute the sound quality optimization operation of the Bluetooth speaker.

9. A computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that, When the processor executes the computer program, the steps of the method for improving the sound quality of the Bluetooth speaker according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method for improving the sound quality of the Bluetooth speaker according to any one of claims 1 to 7 are implemented.

Images