Intelligent control method and system for wireless Bluetooth headset

Abstract

The invention belongs to the technical field of wireless Bluetooth earphones, and particularly relates to an intelligent control method and system for a wireless Bluetooth earphone, and the method comprises a multi-mode instruction interaction method which achieves two convenient interaction control modes of touch control and voice intelligent control through an intelligent touch screen and a voice signal processing system. The intelligent self-adaptive noise reduction method depends on an intelligent voice noise reduction system, noise is automatically filtered by combining an active self-adaptive intelligent noise reduction method, a passive noise reduction method and a call noise reduction method, and external noise interference is reduced. The voice encryption secure communication method is built in a data encryption transmission link, encryption protection is carried out on call data through a Hash algorithm, the call data are prevented from being tampered or interfered, and the integrity and security of the call data are ensured; the Bluetooth data communication system adopts a Bluetooth 3.0 communication protocol, and Bluetooth data transmission with strong stability and high reliability is realized through Wi-Fi data transmission and Gaussian frequency shift keying (GFSK) modulation.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless Bluetooth headsets, and in particular relates to an intelligent control method and system for wireless Bluetooth headsets. Background Art

[0002] With the rapid development of science and technology and the improvement of people's living standards, wireless Bluetooth headsets have become an indispensable part of daily life. They not only provide us with a convenient audio experience, but also continue to integrate new intelligent control technologies to meet the growing and diverse needs of users. The development of wireless Bluetooth headsets can be traced back to the end of the last century. Early wireless Bluetooth headsets mainly used Bluetooth technology for audio transmission, which solved the constraints and inconveniences brought by traditional wired headsets. With the continuous advancement of technology, the sound quality, stability, and battery life of wireless Bluetooth headsets have been significantly improved, and they have gradually become mainstream products on the market. In recent years, with the rapid development of technologies such as the Internet of Things and artificial intelligence, wireless Bluetooth headsets have also ushered in a wave of intelligent innovation. The integration of intelligent control technology has enabled wireless Bluetooth headsets to have more functions, such as voice assistants, health monitoring, environmental perception, etc., bringing users a more convenient and intelligent use experience.

[0003] As consumers become more dependent on smart devices, the demand for smart wireless Bluetooth headsets is also increasing. Users hope that headsets can have more intelligent functions, such as automatically identifying user intentions, providing personalized audio experience, and seamless connection with other smart devices. These demands have promoted the continuous development of intelligent control technology for wireless Bluetooth headsets. At the same time, the rapid development of technologies such as the Internet of Things and artificial intelligence has provided strong support for the development of intelligent control technology for wireless Bluetooth headsets. Internet of Things technology enables headsets to interconnect with other smart devices to achieve data sharing and collaborative work; artificial intelligence technology enables headsets to understand user intentions more accurately and provide more intelligent services through technologies such as machine learning and natural language processing. Finally, fierce market competition has also led manufacturers to increase their investment in the research and development of intelligent control technology for wireless Bluetooth headsets to enhance the competitiveness and market share of their products. This competitive pressure has prompted manufacturers to continuously explore new technical routes and solutions, promoting the continuous advancement of intelligent control technology for wireless Bluetooth headsets.

[0004] In general, although the intelligent control technology of wireless Bluetooth headsets has made significant progress, there are still some defects and shortcomings in practical applications, including: 1. Voice recognition accuracy issues: Although voice recognition technology has made great progress, the voice recognition accuracy of wireless Bluetooth headsets will still be affected to a certain extent in complex environments or when the user's pronunciation is inaccurate. This may result in the user's instructions not being correctly recognized and executed, reducing the user experience.

[0005] 2. Connection stability issues: Since wireless Bluetooth headsets use wireless transmission technology, their connection stability is affected by many factors, such as signal interference, device distance, etc. In an environment with a lot of signal interference, the connection of wireless Bluetooth headsets may become unstable, resulting in audio interruption or delay. This brings troubles to the user experience, especially in situations where real-time audio is required, such as playing games or watching videos.

[0006] 3. Power consumption and battery life: The addition of smart control functions usually leads to an increase in power consumption, which poses a challenge to the battery life of wireless Bluetooth headsets. Due to the size and wearing comfort, the battery capacity of wireless Bluetooth headsets is limited. Long-term use or frequent use of smart control functions may cause rapid power consumption and require frequent charging.

[0007] 4. Compatibility issues: Since different brands and models of wireless Bluetooth headsets may use different intelligent control technologies and standards, this may cause compatibility issues with certain devices or applications. Users need to consider the compatibility of wireless Bluetooth headsets when choosing them to avoid inconvenience caused by compatibility issues.

[0008] 5. Functional redundancy and complexity: With the continuous increase of intelligent control functions, the operation interface and function settings of wireless Bluetooth headsets may become complicated and redundant. This increases the user's learning cost and difficulty of use, and reduces the user's efficiency and satisfaction.

[0009] 6. Data security and privacy protection: Wireless Bluetooth headsets have potential security risks in collecting user data. How to protect the security and privacy of user data and prevent data leakage and abuse are important issues that need to be solved in the intelligent control technology of wireless Bluetooth headsets.

[0010] In summary, there are still some defects and deficiencies in the practical application of wireless Bluetooth headset intelligent control technology. In order to improve user experience and meet user needs, it is necessary to continuously improve and optimize the implementation and performance of intelligent control technology, and at the same time strengthen research and measures in data security and privacy protection to achieve more intelligent and personalized development.

[0011] The prior art CN117579974A discloses a helmet headset with a noise reduction function, including a helmet body and a bone conduction wireless headset. A sound transmission mechanism is provided on one side of the bone conduction wireless headset for transmitting audio information received by a wireless transceiver to the space inside the helmet. A noise reduction adjustment unit is provided on the other side of the bone conduction wireless headset for stably adjusting the spatial position of the bone conduction vibrator in the space inside the helmet and adjusting the external environmental noise transmitted to the space inside the helmet through the bone conduction wireless headset during riding. When riding at high speed, the wind resistance of the wind resistance adjustment plate of the present invention increases, driving the wind resistance adjustment plate to rotate in the opposite direction of traveling, reducing the positive contact area between the wind resistance adjustment plate and the oncoming air, thereby reducing wind noise, and at the same time driving the adjustment magnet close to the repelling magnet, so that the temporal bone patch increases the pressure on the skull, increases the vibration energy received by the skull, and improves the sound quality of audio received by the human ear. However, this invention is only applicable to high-speed riding or scenes with high wind speeds, and mainly uses mechanical noise reduction to reduce the impact of wind noise. It has no noise reduction capability for human and equipment noise interference in daily environment, and because only mechanical noise reduction devices are used, the degree of intelligence and controllability are low. In the prior art CN117476031A, a method and system for enhancing the voice of headset calls in a noisy environment are disclosed, including extracting a pre-selected usage scenario by obtaining feedback information from the headset wearing user, clustering the historical call voice sequence according to the usage scenario, extracting the user voice features, and constructing a voice recognition enhancement model according to the voice features and preset acoustic features; obtaining a call voice sequence containing environmental noise in the current usage scenario, separating the voice subsequence of the user wearing the call voice sequence; obtaining the noise distribution in the voice subsequence, calling the noise attenuation and filtering corresponding to the current usage scenario, and using linear coding to perform linear prediction on the filtered voice subsequence to obtain a continuous call voice sequence. The present invention performs voice gain on the sound field environment in different scenarios, greatly improves the headset call quality, and considers the user voice features of the headset wearing user, provides a personalized noise reduction effect, and improves the call comfort of the headset wearing user. However, this technology mainly realizes the user voice enhancement during the user call process, reduces the noise interference caused by non-user voice during the call process, and the application scenario is relatively limited and single. In non-voice call scenarios, such as watching movies and playing music on mobile phones, the surrounding environment noise cannot be effectively eliminated and reduced. To this end, the present invention provides an intelligent control method and system for a wireless Bluetooth headset. Summary of the invention

[0012] In order to solve the above problems, the present invention proposes an intelligent control method and system for a wireless Bluetooth headset to solve the problems existing in the above-mentioned prior art, improve the effectiveness and efficiency of intelligent control of wireless Bluetooth headsets, and improve user experience.

[0013] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent control method for wireless Bluetooth headsets, applied to the intelligent control system of wireless Bluetooth headsets, including a multi-mode command interaction method, an intelligent adaptive noise reduction method, and a voice encryption and secure communication method.

[0014] Furthermore, in the intelligent control method of a wireless Bluetooth headset, the multi-mode command interaction method includes two methods: touch control and voice control.

[0015] Furthermore, the intelligent control method of a wireless Bluetooth headset, the touch control method includes installing a touch sensing area on the surface of the wireless headset, executing different control instructions according to different touch, press, and tapping rhythms of fingers, and users can use the system default tapping instructions or independently set tapping instructions through the connected smartphone.

[0016] Furthermore, the intelligent control method of the wireless Bluetooth headset, the voice control method has built-in intelligent voice recognition, semantic text decoding and natural language processing algorithms, which can accurately recognize and understand user semantic instructions and accurately control the Bluetooth headset, and the specific steps include: Step 1: Construct a voice database: Collect voice data from people of different countries, regions, genders, and age groups through recording methods. The voice data includes various official languages ​​and regional dialects with different tones and timbres, and initially form a complete and comprehensive voice database with corresponding text data as labels.

[0017] Step 2: Data preprocessing: Preprocess the speech data and text labels in the initially constructed speech database. The specific preprocessing operations include: (1) Data denoising The environmental noise in the voice data is filtered based on the wavelet transform algorithm to improve the signal-to-noise ratio of the voice data.

[0018] (2) Endpoint Detection Find the starting and ending points of the speech signal to help determine the effective part of the speech signal and avoid unnecessary processing of the silent or noisy parts.

[0019] (3) Pre-emphasis The high-frequency part of the speech is emphasized through a high-pass digital filter to remove the influence of lip radiation and increase the high-frequency resolution of the speech. The pre-emphasis coefficient range is set between 0.9 and 1.0.

[0020] (4) Framing Since the speech signal is not stable as a whole, it needs to be divided into shorter frames for processing. Each frame contains 10 to 30 milliseconds of speech signal, and the ratio of frame shift to frame length ranges from 0 to 1 / 2.

[0021] (5) Feature extraction The short-time energy, short-time average zero-crossing rate, linear prediction coefficient and Mel-frequency cepstral coefficient features are extracted from the speech signal to model and analyze the speech signal.

[0022] (6) Dimensionality reduction The linear discriminant analysis method is used to reduce the dimension of speech features and reduce the impact of excessive feature dimension on model training.

[0023] (7) Standardization The mean normalization method is used to standardize the speech, making the speech of different people closer in characteristics and improving the generalization ability of the model.

[0024] (8) Volume normalization The volume of the speech signal is adjusted by the constant amplitude method to make it as stable as possible, thereby improving the analyzability and recognizability of the speech signal.

[0025] Step 3: Neural network model training: Build a neural network model, divide the voice data processed in step 2 into a training data set and a test data set in an 8:2 ratio, and train the neural network model on the training data set, using voice data as input and the corresponding text label as the optimization target. Monitor the training process. When the loss curve tends to be smooth and no longer drops significantly, the model is considered to have completed model training.

[0026] Step 4: Intelligent speech recognition: Load the model weights and test the model performance on the test dataset.

[0027] Step 5: Semantic text decoding: Deploy the neural network that has passed the performance test into the system to implement a speech recognition method that maps and converts speech signals into text signals.

[0028] Step 6: Natural language processing: Build a neural network model for natural language processing tasks, parse and semantically understand the corresponding text information generated after speech recognition and semantic text decoding, accurately understand user voice commands based on the processed results, and execute related command tasks.

[0029] Furthermore, the intelligent control method of the wireless Bluetooth headset, the intelligent adaptive noise reduction method includes an active adaptive intelligent noise reduction method, a passive noise reduction method and a call noise reduction method.

[0030] Furthermore, the intelligent control method of the wireless Bluetooth headset and the active adaptive intelligent noise reduction method specifically include: Step 1: Construct sound signal datasets of noisy human voice and clean human voice respectively, where the noisy human voice data is used as the input of the neural network, and the clean human voice is used as the label corresponding to the noisy human voice data, which is also the optimization target of the neural network.

[0031] Step 2: Change the noise intensity, construct noisy data and corresponding labels under noise interference of different intensities, and form a sound signal denoising dataset to improve the feature extraction and generalization capabilities of the neural network model.

[0032] Step 3: Construct a two-stage sound signal denoising neural network model, and complete the training and testing of the neural network model on the data set constructed in step 2.

[0033] Step 4: The microphone on the wireless Bluetooth headset collects noise signals from the surrounding environment. The noise signal processor inside the headset will reversely process the collected noise signals to generate a sound wave signal with a phase difference of 180 degrees and roughly the same amplitude as the noise.

[0034] Step 5: The generated sound wave signal that is opposite to the noise is mixed with the audio signal and noise signal played by the earphone and sent to the output system of the earphone to cancel out part of the noise.

[0035] Step 6: Send the mixed sound signal after the offset processing in step 5 to the neural network model after training and testing in step 3 to achieve further refined elimination of the noise signal.

[0036] Step 7: The sound signal after the two noise reduction processes is output to the human ear through the speaker of the headset, thereby reducing the interference of external noise to the user.

[0037] Furthermore, the intelligent control method of the wireless Bluetooth headset and the passive noise reduction method are to set the wireless Bluetooth headset as an in-ear headset, and to fit the earplug tightly with the ear canal of the human ear to form a relatively closed environment, and to isolate external noise through a physical noise reduction method to reduce noise interference.

[0038] Furthermore, the intelligent control method of the wireless Bluetooth headset and the call noise reduction method utilize the noise reduction chip inside the wireless Bluetooth headset to filter the signal received by the call microphone to reduce external interference such as wind noise, and at the same time, the echo of the call sound transmitted from the other party can be weakened to improve the user's call quality.

[0039] Furthermore, the intelligent control method of the wireless Bluetooth headset and the voice encryption and security communication method adopt a hash algorithm to protect call data, prevent the call data from being tampered with or interfered with, and ensure the integrity and security of the call data.

[0040] On the other hand, an intelligent control system of a wireless Bluetooth headset is provided and applied to any one of the intelligent control methods of a wireless Bluetooth headset, wherein the intelligent control system of the wireless Bluetooth headset comprises: Smart touch screen: The smart touch screen adopts pressure-sensitive touch technology and has a built-in capacitor plate. The user's finger is used as the other plate of the capacitor. When the user presses or taps the smart touch screen, the distance between the two plates of the capacitor changes, and the on-off state of the current also changes. Different built-in instructions correspond to different user tapping rhythms, and different operation instructions are executed. At the same time, the built-in fingerprint recognition system can intelligently identify the user's identity based on the fingerprint information entered by the connected mobile phone to ensure the safety of the user's wireless Bluetooth headset.

[0041] Bluetooth data communication system: The Bluetooth data communication system is used to realize data interaction between users, mobile phones and wireless Bluetooth headsets. It adopts Bluetooth 3.0 communication protocol and transmits data through Wi-Fi. The transmission rate is greater than 1Mb / s. The modulation method is Gaussian frequency shift keying modulation, and the modulation index range is 0.28-0.35. A Gaussian filter is used to smooth the frequency switching process of the modulated signal, convert the digital signal into a continuous sinusoidal wave signal, and transmit information by changing the offset of the carrier frequency, thereby improving the signal's spectrum efficiency, anti-interference ability and anti-frequency deviation ability.

[0042] Data encryption transmission link: The data encryption transmission link encrypts and protects user voice data, text data and other personal data through encryption algorithms to avoid external noise, malicious interference and personal privacy leakage.

[0043] Voice signal processing system: The voice signal processing system consists of two parts: a voice signal acquisition subsystem and a voice signal intelligent recognition subsystem. The voice signal acquisition subsystem includes a single microphone, which is responsible for collecting the user's voice signal; the voice signal intelligent recognition subsystem has a built-in neural network model that can effectively recognize and understand the user's voice signal and execute related instructions.

[0044] Intelligent voice noise reduction system: The intelligent voice noise reduction system consists of a noise signal processor, an intelligent noise reduction subsystem and a noise reduction chip. The noise signal processor is located at the back end of the microphone, analyzes the collected noise signal, gives parameter information such as noise signal power, amplitude, phase, and generates a sound wave signal with a phase difference of 180 degrees and roughly the same amplitude as the noise to offset part of the noise interference; the intelligent noise reduction subsystem contains a corresponding neural network model to further refine the noise reduction processing of the sound signal after the initial noise reduction by the noise signal processor; the noise reduction chip inside the wireless Bluetooth headset filters the signal received by the call microphone to reduce external interference such as wind noise, and can also perform echo weakening processing on the call sound transmitted by the other party to improve the user's call quality.

[0045] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention realizes effective control of wireless Bluetooth headsets through two modes: press touch and intelligent voice recognition control, which facilitates users to freely choose the control mode according to their current state, and provides users with more convenient, intelligent and more diversified wireless Bluetooth headset control services.

[0046] 2. The present invention realizes intelligent, reliable and efficient voice noise reduction in different scenarios and user needs through a joint noise reduction method of active adaptive noise reduction, passive noise reduction and call noise reduction, supports intelligent noise reduction services in multiple modes such as voice calls, audio playback, video entertainment, etc., helps to improve the service quality of wireless Bluetooth headsets in multiple scenarios and improve user experience.

[0047] 3. The present invention improves the security of user voice data and data interaction between wireless Bluetooth headsets and user's personal mobile terminals through a voice encryption security communication method, avoids personal data information leakage caused by malicious eavesdropping and noise interference, and protects user's personal privacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flow chart of the intelligent control method of the wireless Bluetooth headset of the present invention; Figure 2 This is a structural diagram of the wireless Bluetooth headset intelligent control system of the present invention; DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0050] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0051] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0052] In an embodiment of the present invention, a method for intelligently controlling a wireless Bluetooth headset is provided. Figure 1 , applied to an intelligent control system of a wireless Bluetooth headset, including a multi-mode command interaction method, an intelligent adaptive noise reduction method, and a voice encryption and secure communication method.

[0053] The multi-mode command interaction methods include touch control and voice control. The touch control method controls the touch sensing area installed on the surface of the wireless headset through operations such as touching, pressing, and tapping rhythm, thereby realizing different control commands; the voice control method has built-in intelligent voice recognition, semantic text decoding, and natural language processing algorithms, which can accurately identify and understand user semantic commands, and perform accurate control of the Bluetooth headset.

[0054] Intelligent adaptive noise reduction methods include active adaptive intelligent noise reduction methods, passive noise reduction methods and call noise reduction methods. The active adaptive intelligent noise reduction method combines a noise signal processor and a two-stage intelligent adaptive denoising neural network algorithm. The noise processor reversely processes the collected noise signal to generate a sound wave signal that is 180 degrees out of phase with the noise and has approximately the same amplitude, and preliminarily cancels and removes the noise signal. The two-stage intelligent adaptive denoising neural network algorithm achieves further refined elimination of the noise signal. The passive noise reduction method designs the physical structure of the wireless Bluetooth headset as an in-ear headset, and fits the earplug tightly with the ear canal of the human ear to form a relatively closed environment, and isolates external noise through physical noise reduction methods to reduce noise interference. The call noise reduction method uses the noise reduction chip inside the wireless Bluetooth headset to filter the signal received by the call microphone to reduce external interference such as wind noise. At the same time, it can also perform echo weakening on the call sound transmitted from the other party to improve the user's call quality.

[0055] The voice encryption secure communication method uses a hash algorithm to protect call data, prevent call data from being tampered with or interfered with, and ensure the integrity and security of call data.

[0056] Refer to the attached Figure 2 The present invention also provides an intelligent control system for a wireless Bluetooth headset, including an intelligent touch screen, a Bluetooth data communication system, a data encryption transmission link, a voice signal processing system and an intelligent voice noise reduction system.

[0057] The smart touch screen uses pressure-sensitive touch technology and has a built-in capacitor plate. The user's finger is used as the other plate of the capacitor. When the user presses or taps the smart touch screen, the distance between the two capacitor plates changes, and the current on and off state also changes. Different built-in instructions correspond to different user tapping rhythms, and different operation instructions are executed. At the same time, the built-in fingerprint recognition system can intelligently identify the user's identity based on the fingerprint information entered by the connected mobile phone, ensuring the safety of the user's wireless Bluetooth headset use.

[0058] The Bluetooth data communication system is used to realize data interaction between users, mobile phones and wireless Bluetooth headsets. It adopts Bluetooth 3.0 communication protocol and transmits data through Wi-Fi with a transmission rate greater than 1Mb / s. The modulation method is Gaussian frequency shift keying (GFSK) modulation with a modulation index range of 0.28-0.35. A Gaussian filter is used to smooth the frequency switching process of the modulated signal, convert the digital signal into a continuous sinusoidal wave signal, and transmit information by changing the offset of the carrier frequency, thereby improving the signal's spectrum efficiency, anti-interference ability and anti-frequency deviation ability.

[0059] The data encryption transmission link encrypts and protects user voice data, text data and other personal data through encryption algorithms to avoid external noise, malicious interference and personal privacy leakage.

[0060] The voice signal processing system consists of two parts: the voice signal acquisition subsystem and the voice signal intelligent recognition subsystem. The voice signal acquisition subsystem contains a single microphone, which is responsible for collecting the user's voice signal; the voice signal intelligent recognition subsystem has a built-in neural network model that can effectively recognize and understand the user's voice signals and execute related instructions.

[0061] The intelligent voice noise reduction system consists of a noise signal processor, an intelligent noise reduction subsystem and a noise reduction chip. The noise signal processor is located at the back end of the microphone and analyzes the collected noise signal to provide parameter information such as noise signal power, amplitude, phase, etc. It also generates a sound wave signal with a phase difference of 180 degrees and roughly the same amplitude as the noise to offset some noise interference; the intelligent noise reduction subsystem contains a corresponding neural network model to further refine the noise reduction processing of the sound signal after the initial noise reduction by the noise signal processor; the noise reduction chip inside the wireless Bluetooth headset filters the signal received by the call microphone to reduce external interference such as wind noise. At the same time, it can also perform echo weakening processing on the call sound transmitted by the other party to improve the user's call quality.

[0062] In a specific embodiment, the user of the wireless Bluetooth headset realizes direct control of the wireless Bluetooth headset through a multi-mode command interaction method. In conventional usage scenarios, the user can set touch control instructions corresponding to different control instructions in the mobile phone or other personal terminal connected to the Bluetooth headset according to personal preferences and usage habits in advance. Through the built-in pressure-sensitive touch technology and touch control method of the smart touch screen, the common quick control of the wireless Bluetooth headset such as answering calls, hanging up calls, playing music, adjusting volume, and pausing music operations is realized by using a combination of touch, pressing, and tapping rhythm. When the user is in sports mode or both hands are occupied, the wireless Bluetooth headset can be intelligently controlled by voice through the intelligent voice recognition method built into the voice signal processing system. The Bluetooth data communication system used by the wireless Bluetooth headset can provide stable real-time information transmission in a strong interference environment, and realize stable communication over a long distance between the wireless Bluetooth headset and the mobile phone terminal with extremely low latency. When the user is in a noisy environment, the wireless Bluetooth headset intelligent voice noise reduction system will automatically start working, accurately, thoroughly, and effectively filter out the external environmental noise interference, ensuring that the user is not disturbed by external noise when answering calls, enjoying music, and watching videos, thereby improving the user experience. When users need to use wireless Bluetooth headsets for important internal meetings or personal private calls, the built-in voice encryption and security communication method of the wireless Bluetooth headset control system uses a hash algorithm to protect call data, prevent call data from being tampered with or interfered with, and ensure the integrity and security of call data.

[0063] Example 1

[0064] In one embodiment, in a conventional usage scenario, a user of a wireless Bluetooth headset implements conventional and convenient control operations on the wireless Bluetooth headset through a touch control method in a multi-mode command interaction method. The specific steps include: Step 1: The wireless Bluetooth headset user opens the Bluetooth connection module of the mobile phone, selects the name of the wireless Bluetooth headset on the mobile phone Bluetooth connection interface, and completes the pairing of the wireless Bluetooth headset; Step 2: After pairing is completed, the user can open the wireless Bluetooth headset settings page on the mobile phone interface and set and modify convenient control commands according to personal usage habits and preferences. For example, tap the smart touch screen once to pause and continue playing music, tap twice in a row to answer and interrupt calls, rotate clockwise to increase the volume, and rotate counterclockwise to reduce the volume. Step 3: After completing the settings and modifications, click the Save button on the interface and wait for the Bluetooth data communication system to synchronize data before realizing touch intelligent control.

[0065] When the user is in motion, has both hands occupied, or wants to achieve more command control, the intelligent voice control method built into the voice signal processing system can be used to achieve intelligent control of the wireless Bluetooth headset. The specific steps include: Step 1: Construct a voice database: Collect voice data from people of different countries, regions, genders, and age groups through recording methods. The voice data includes various official languages ​​and regional dialects with different tones and timbres, and initially form a complete and comprehensive voice database with corresponding text data as labels.

[0066] Step 2: Data preprocessing: Preprocess the speech data and text labels in the initially constructed speech database. The specific preprocessing operations include: (1) Data denoising The environmental noise in the voice data is filtered based on the wavelet transform algorithm to improve the signal-to-noise ratio of the voice data. The specific steps include: 1) Wavelet decomposition of noisy speech signals: Select a suitable wavelet basis and determine the number of decomposition layers l, perform decomposition calculations, and obtain the high-frequency coefficients of each layer of wavelet decomposition .

[0067] 2) Threshold quantization of high-frequency coefficients. Choose an appropriate threshold and threshold function Threshold processing is performed on the high-frequency coefficients of each layer In the threshold quantization process, two functions can be used: soft threshold function and hard threshold function. The expression of hard threshold function is: ; The soft threshold function expression is: ; in, is the wavelet coefficient, is the wavelet coefficient after threshold processing, For the j Threshold for high frequency coefficients of a layer.

[0068] (2) Endpoint Detection Find the starting and ending points of the speech signal to help determine the effective part of the speech signal and avoid unnecessary processing of the silent or noisy parts.

[0069] (3) Pre-emphasis The high-frequency part of the speech is emphasized through a high-pass digital filter to remove the influence of lip radiation and increase the high-frequency resolution of the speech. The pre-emphasis coefficient range is set between 0.9 and 1.0. The high-pass filter adopts an active high-pass filter, which is composed of passive components and an integrated operational amplifier. The signal in the passband not only has no energy loss, but also can be amplified, and the load effect is not obvious. When multiple stages are connected, the mutual influence is very small. It is easy to form a high-order filter using a simple cascade method, and the filter is small in size, light in weight, and does not require magnetic shielding.

[0070] (4) Framing Since the speech signal is not stable as a whole, it needs to be divided into shorter frames for processing. Each frame contains 10 to 30 milliseconds of speech signal, and the ratio of frame shift to frame length ranges from 0 to 1 / 2.

[0071] (5) Feature extraction The short-time energy, short-time average zero-crossing rate, linear prediction coefficient and Mel-frequency cepstral coefficient features are extracted from the speech signal to model and analyze the speech signal.

[0072] (6) Dimensionality reduction The linear discriminant analysis (LDA) method is used to reduce the dimension of speech features and reduce the impact of high feature dimensions on model training. Linear discriminant analysis is a supervised learning algorithm whose main goal is to provide effective data representation for classification problems by maximizing the difference between categories while reducing the dimension. The specific steps include: 1) Define the intra-category scatter matrix of each of the above category features as:

[0073] This formula reflects the degree of discreteness of data within the same category; 2) Define the inter-category scatter matrix of the data between the above different category features: ; This formula reflects the discrete degree of data between different categories, where Respectively represent The collection and mean vector of class examples.

[0074] 3) The optimization goal of the algorithm is to make the projection points of similar samples as close as possible, that is, As small as possible, and the projection points of heterogeneous samples are as far away as possible, that is, As large as possible. Considering both at the same time, we can get the maximum optimization goal as: ; From the above formula, we can see that The solution and The length is irrelevant, only is related to the direction, so we can make , it can be equivalently converted to: ; By Lagrange multiplier method, the above formula is equivalent to: ;in is the Lagrange multiplier method, and from the above derivation we can get: ; and because is a scalar quantity, which depends only on the direction, so we can make , combining the above formulas, we can get: ;make ,have

[0075] 4) The data that will eventually be projected onto the straight line The feature is sent to a fixed threshold classifier to determine the category.

[0076] (7) Standardization The mean normalization method is used to standardize the speech, making the speech of different people closer in characteristics and improving the generalization ability of the model.

[0077] (8) Volume normalization The volume of the speech signal is adjusted by the constant amplitude method to make it as stable as possible, thereby improving the analyzability and recognizability of the speech signal.

[0078] Step 3: Neural network model training: Since the speech signal is a time-continuous data stream signal, a bidirectional long-short time neural network model is constructed. The speech data processed in step 2 is divided into a training data set and a test data set in an 8:2 ratio. The neural network model is trained on the training data set, with speech data as input and the corresponding text label as the optimization target. The training process is monitored. When the loss curve tends to be smooth and no longer drops significantly, the model is considered to have completed model training.

[0079] Step 4: Intelligent speech recognition: Load the model weights and test the model performance on the test dataset.

[0080] Step 5: Semantic text decoding: Deploy the neural network that has passed the performance test into the system to implement a speech recognition method that maps and converts speech signals into text signals.

[0081] Step 6: Natural language processing: Build a Transformer neural network model for natural language processing tasks, parse and semantically understand the corresponding text information generated after speech recognition and semantic text decoding, accurately understand user voice commands based on the processed results, and execute related command tasks.

[0082] Example 2 In one embodiment, the wireless Bluetooth headset manufacturer improves the noise reduction performance of the wireless Bluetooth headset and improves the user experience through the intelligent adaptive noise reduction method built into the intelligent voice noise reduction system. The intelligent adaptive noise reduction method includes an active adaptive intelligent noise reduction method, a passive noise reduction method and a call noise reduction method. The passive noise reduction method designs the physical structure of the wireless Bluetooth headset as an in-ear headset, and the earplug fits tightly with the ear canal of the human ear to form a relatively closed environment. The physical noise reduction method isolates external noise and reduces noise interference. The call noise reduction method uses the noise reduction chip inside the wireless Bluetooth headset to filter the signal received by the call microphone to reduce external interference such as wind noise. At the same time, the call sound transmitted from the other party can also be echo weakened to improve the user's call quality. The specific steps of the active adaptive intelligent noise reduction method include: Step 1: Construct sound signal datasets of noisy human voice and clean human voice respectively, where the noisy human voice data is used as the input of the neural network, and the clean human voice is used as the label corresponding to the noisy human voice data, which is also the optimization target of the neural network.

[0083] Step 2: Change the noise intensity, construct noisy data and corresponding labels under noise interference of different intensities, and form a sound signal denoising dataset to improve the feature extraction and generalization capabilities of the neural network model.

[0084] Step 3: Construct a two-stage sound signal denoising neural network model CBDNet. The neural network includes a noise estimation subnetwork and a denoising subnetwork. The noise estimation subnetwork is responsible for estimating the noise intensity in the sound signal, including information such as noise amplitude, phase, and signal-to-noise ratio. The denoising subnetwork performs accurate noise removal based on the noise information estimated by the noise estimation subnetwork, and completes the training and testing of the neural network model on the data set constructed in step 2.

[0085] Step 4: The microphone on the wireless Bluetooth headset collects noise signals from the surrounding environment. The noise signal processor inside the headset will reversely process the collected noise signals to generate a sound wave signal with a phase difference of 180 degrees and roughly the same amplitude as the noise.

[0086] Step 5: The generated sound wave signal that is opposite to the noise is mixed with the audio signal and noise signal played by the earphone and sent to the output system of the earphone to cancel out part of the noise.

[0087] Step 6: Send the mixed sound signal after the offset processing in step 5 to the neural network model after training and testing in step 3 to achieve further refined elimination of the noise signal.

[0088] Step 7: Output the sound signal after two noise reduction processes to the human ear through the earphone speaker, thereby reducing the interference of external noise to the user.

[0089] Embodiment 3 In one embodiment, the wireless Bluetooth headset manufacturer improves the security and privacy of wireless Bluetooth headset communication through a voice encryption secure communication method built into the data encryption transmission link. The voice encryption secure communication method adopts the MD5 algorithm in the hash function. The specific steps include: (1) Padding: Output the length of the input voice signal in bits, and divide this length by 512 to take the remainder. If the remainder is not equal to 448, then it is necessary to pad with a 1 and n zeros such that the remainder of the padded information length divided by 512 is 448. At this time, the length of the information is: N *512 + 448 (bits); (2) Record the information length Use 64 bits to store the information length before padding, and place these 64 bits behind the result of step (1). At this time, the information length is N *512 + 448 = ( N +1) (bits); (3) Load the standard magic numbers (four integers) The standard magic numbers are all hexadecimal numbers. The physical order is: A = (01234567)16, B = (89ABCDEF)16, C = (FEDCBA98)16, D = (76543210)16; (4) Four rounds of cyclic operations 1) Set the maximum number of loops to the number of groups plus 1; 2) Subdivide each 512-byte into 16 cyclic groups, each group with 64 bits, that is, 8 bytes; 3) Define four linear functions F(X,Y,Z)=(X&Y)|((~X)&Z) G(X,Y,Z)=(X&Z)|(Y&(~Z)) H(X,Y,Z)=X^Y^Z I(X,Y,Z)=Y^(X|(~Z)) Among them, & represents the "AND" operation in logical operations, | represents the "OR" operation in logical operations, ~ represents the "NOT" operation in logical operations, and ^ represents the "XOR" operation in logical operations; 4) Let Mj represent the jth sub-group of the message (from 0 to 15), and <<<s represent a cyclic left shift of s bits. Then the four-round operation can be defined as: FF(a, b, c, d, Mj, s, ti) means a = b + ((a + F(b, c, d) + Mj + ti) <<< s) GG(a, b, c, d, Mj, s, ti) means a = b + ((a + G(b, c, d) + Mj + ti) <<< s) HH(a, b, c, d, Mj, s, ti) means a = b + ((a + H(b, c, d) + Mj + ti) <<< s) II(a, b, c, d, Mj, s, ti) means a = b + ((a + I(b, c, d) + Mj + ti) <<< s) 5) Four-round operation The first round a = FF(a, b, c, d, M0, 7, 0xd76aa478) b = FF(d, a, b, c, M1, 12, 0xe8c7b756) c = FF(c, d, a, b, M2, 17, 0x242070db) d = FF(b, c, d, a, M3, 22, 0xc1bdceee) a = FF(a, b, c, d, M4, 7, 0xf57c0faf) b = FF(d, a, b, c, M5, 12, 0x4787c62a) c = FF(c, d, a, b, M6, 17, 0xa8304613) d = FF(b, c, d, a, M7, 22, 0xfd469501) a = FF(a, b, c, d, M8, 7, 0x698098d8) b = FF(d, a, b, c, M9, 12, 0x8b44f7af) c = FF(c, d, a, b, M10, 17, 0xffff5bb1) d = FF(b, c, d, a, M11, 22, 0x895cd7be) a = FF(a, b, c, d, M12, 7, 0x6b901122) b = FF(d, a, b, c, M13, 12, 0xfd987193) c = FF(c, d, a, b, M14, 17, 0xa679438e) d = FF(b, c, d, a, M15, 22, 0x49b40821) The second round a = GG(a, b, c, d, M1, 5, 0xf61e2562) b = GG(d, a, b, c, M6, 9, 0xc040b340) c = GG(c, d, a, b, M11, 14, 0x265e5a51) d = GG(b, c, d, a, M0, 20, 0xe9b6c7aa) a = GG(a, b, c, d, M5, 5, 0xd62f105d) b = GG(d, a, b, c, M10, 9, 0x02441453) c = GG(c, d, a, b, M15, 14, 0xd8a1e681) d = GG(b, c, d, a, M4, 20, 0xe7d3fbc8) a = GG(a, b, c, d, M9, 5, 0x21e1cde6) b = GG(d, a, b, c, M14, 9, 0xc33707d6) c = GG(c, d, a, b, M3, 14, 0xf4d50d87) d = GG(b, c, d, a, M8, 20, 0x455a14ed) a = GG(a, b, c, d, M13, 5, 0xa9e3e905) b = GG(d, a, b, c, M2, 9, 0xfcefa3f8) c = GG(c, d, a, b, M7, 14, 0x676f02d9) d = GG(b, c, d, a, M12, 20, 0x8d2a4c8a) The third round a = HH(a, b, c, d, M5, 4, 0xfffa3942) b = HH(d, a, b, c, M8, 11, 0x8771f681) c = HH(c, d, a, b, M11, 16, 0x6d9d6122) d = HH(b, c, d, a, M14, 23, 0xfde5380c) a = HH(a, b, c, d, M1, 4, 0xa4beea44) b = HH(d, a, b, c, M4, 11, 0x4bdecfa9) c = HH(c, d, a, b, M7, 16, 0xf6bb4b60) d = HH(b, c, d, a, M10, 23, 0xbebfbc70) a = HH(a, b, c, d, M13, 4, 0x289b7ec6) b = HH(d, a, b, c, M0, 11, 0xeaa127fa) c = HH(c, d, a, b, M3, 16, 0xd4ef3085) d = HH(b, c, d, a, M6, 23, 0x04881d05) a = HH(a, b, c, d, M9, 4, 0xd9d4d039) b = HH(d, a, b, c, M12, 11, 0xe6db99e5) c = HH(c, d, a, b, M15, 16, 0x1fa27cf8) d = HH(b, c, d, a, M2, 23, 0xc4ac5665) Fourth round a = II(a, b, c, d, M0, 6, 0xf4292244) b = II(d, a, b, c, M7, 10, 0x432aff97) c = II(c, d, a, b, M14, 15, 0xab9423a7) d = II(b, c, d, a, M5, 21, 0xfc93a039) a = II(a, b, c, d, M12, 6, 0x655b59c3) b = II(d, a, b, c, M3, 10, 0x8f0ccc92) c = II(c, d, a, b, M10, 15, 0xffeff47d) d = II(b, c, d, a, M1, 21, 0x85845dd1) a = II(a, b, c, d, M8, 6, 0x6fa87e4f) b = II(d, a, b, c, M15, 10, 0xfe2ce6e0) c = II(c, d, a, b, M6, 15, 0xa3014314) d = II(b, c, d, a, M13, 21, 0x4e0811a1) a = II(a, b, c, d, M4, 6, 0xf7537e82) b = II(d, a, b, c, M11, 10, 0xbd3af235) c=II(c,d,a,b,M2,15,0x2ad7d2bb) d=II(b,c,d,a,M9,21,0xeb86d391) 6) After each cycle, add A, B, C, and D to a, b, c, and d respectively, and then enter the next cycle; 7) The MD5 algorithm performs MD5 verification on the user data sent by the Bluetooth headset and the data received by the mobile terminal to prevent the user data from being maliciously tampered with; 8) The user password in the mobile phone system is stored in the file system after MD5 hash operation. When the user logs in, the system performs MD5 hash operation on the password entered by the user, and then compares it with the MD5 value stored in the file system to determine whether the entered password is correct. Through such steps, the system can determine the legitimacy of the user's login to the system without knowing the plain text of the user's password, thereby protecting the user's personal information security.

[0090] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0091] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent control method for a wireless Bluetooth headset, comprising a multi-mode command interaction method, an intelligent adaptive noise reduction method, and a voice encryption secure communication method, characterized in that The multi-mode instruction interaction method includes a touch control method and a voice control method; the intelligent adaptive noise reduction method includes an active adaptive intelligent noise reduction method, a passive noise reduction method and a call noise reduction method; The voice encryption secure communication method uses a hash algorithm to protect call data, prevent call data from being tampered with or interfered with, and ensure the integrity and security of call data.

2. The intelligent control method of a wireless Bluetooth headset according to claim 1, characterized in that: The touch control method specifically comprises the following steps: Step 1: The wireless Bluetooth headset user opens the Bluetooth connection module of the mobile phone, selects the name of the wireless Bluetooth headset on the mobile phone Bluetooth connection interface, and completes the pairing of the wireless Bluetooth headset; Step 2: After pairing is completed, the user can open the wireless Bluetooth headset settings page on the mobile phone interface and set and modify convenient control commands according to personal usage habits and preferences. For example, tap the smart touch screen once to pause and continue playing music, tap twice in a row to answer and interrupt calls, rotate clockwise to increase the volume, and rotate counterclockwise to reduce the volume. Step 3: After completing the settings and modifications, click the Save button on the interface and wait for the Bluetooth data communication system to synchronize data before realizing touch intelligent control.

3. The intelligent control method of a wireless Bluetooth headset according to claim 1, characterized in that: The specific steps of the voice control method include: Step 1: Construct a voice database: Collect voice data from people of different countries, regions, genders, and age groups through recording methods. The voice data includes various official languages ​​and regional dialects with different tones and timbres, and initially form a complete and comprehensive voice database, with corresponding text data as labels; Step 2: Data preprocessing: Preprocess the speech data and text labels in the initially constructed speech database. The specific preprocessing operations include: (1) Data denoising The environmental noise in the voice data is filtered based on the wavelet transform algorithm to improve the signal-to-noise ratio of the voice data. The specific steps include: 1) Wavelet decomposition of noisy speech signals: Select a suitable wavelet basis and determine the number of decomposition layers l, perform decomposition calculations, and obtain the high-frequency coefficients of each layer of wavelet decomposition and ; 2) Threshold quantization of high-frequency coefficients; choosing an appropriate threshold and threshold function Threshold processing is performed on the high-frequency coefficients of each layer In the threshold quantization process, two functions can be used: soft threshold function and hard threshold function. The expression of hard threshold function is: ； The soft threshold function expression is: ； in, is the wavelet coefficient, is the wavelet coefficient after threshold processing, For the j Threshold of high frequency coefficients of the layer; (2) Endpoint Detection Find the starting and ending points of the speech signal to help determine the effective part of the speech signal and avoid unnecessary processing of the silent or noisy parts; (3) Pre-emphasis The high-frequency part of the speech is emphasized through a high-pass digital filter to remove the influence of lip radiation and increase the high-frequency resolution of the speech. The pre-emphasis coefficient range is set between 0.9 and 1.

0. The high-pass filter adopts an active high-pass filter, which is composed of passive components and integrated operational amplifiers. The signal in the passband not only has no energy loss, but also can be amplified, and the load effect is not obvious. When multiple stages are connected, the mutual influence is very small. It is easy to form a high-order filter using a simple cascade method, and the filter is small in size, light in weight, and does not require magnetic shielding; (4) Framing Since the speech signal is not stable as a whole, it needs to be divided into shorter frames for processing. Each frame contains 10 to 30 milliseconds of speech signal, and the ratio of frame shift to frame length ranges from 0 to 1 / 2. (5) Feature extraction Extract the short-time energy, short-time average zero-crossing rate, linear prediction coefficient and Mel-frequency cepstral coefficient features from the speech signal to model and analyze the speech signal; (6) Dimensionality reduction The linear discriminant analysis (LDA) method is used to reduce the dimension of speech features and reduce the impact of high feature dimensions on model training. Linear discriminant analysis is a supervised learning algorithm whose main goal is to provide effective data representation for classification problems by maximizing the difference between categories while reducing the dimension. The specific steps include: 1) Define the intra-category scatter matrix of each of the above category features as: ； This formula reflects the degree of discreteness of data within the same category; 2) Define the inter-category scatter matrix of the data between the above different category features: ； This formula reflects the degree of discreteness of data between different categories, where Respectively represent the set and mean vector of class examples; 3) The optimization goal of the algorithm is to make the projection points of similar samples as close as possible, that is, As small as possible, and the projection points of heterogeneous samples are as far away as possible, that is, As large as possible. Considering both at the same time, we can get the maximum optimization goal as: ， From the above formula, we can see that The solution and The length is irrelevant, only is related to the direction, so we can make , it can be equivalently converted to: ； By Lagrange multiplier method, the above formula is equivalent to: ； in is the Lagrange multiplier method, and from the above derivation we can get: ； Also because is a scalar quantity, which depends only on the direction, so we can make , combining the above formulas, we can get: ； make ,have ； 4) The data that will eventually be projected onto the straight line Send it to a fixed threshold classifier to determine the category of the feature; (7) Standardization The mean normalization method is used to standardize the speech, making the speech of different people more similar in features and improving the generalization ability of the model; (8) Volume normalization The volume of the speech signal is adjusted by the constant amplitude method to make it as stable as possible, thereby improving the analyzability and recognizability of the speech signal. Step 3: Neural network model training: Since the speech signal is a time-continuous data stream signal, a bidirectional long-short time neural network model is constructed. The speech data processed in step 2 is divided into a training data set and a test data set in an 8:2 ratio. The neural network model is trained on the training data set, with the speech data as input and the corresponding text label as the optimization target. The training process is monitored. When the loss curve tends to be smooth and no longer drops significantly, the model is considered to have completed model training. Step 4: Intelligent speech recognition: load the model weights and test the model performance on the test data set; Step 5: Semantic text decoding: Deploy the neural network that has passed the performance test into the system to implement the speech recognition method that maps and converts speech signals into text signals; Step 6: Natural language processing: Build a Transformer neural network model for natural language processing tasks, parse and semantically understand the corresponding text information generated after speech recognition and semantic text decoding, accurately understand user voice commands based on the processed results, and execute related command tasks.

4. The intelligent control method of a wireless Bluetooth headset according to claim 1, characterized in that: The active adaptive intelligent noise reduction method specifically comprises the following steps: Step 1: Construct sound signal datasets of noisy human voice and clean human voice respectively, where the noisy human voice data is used as the input of the neural network, and the clean human voice is used as the label corresponding to the noisy human voice data, which is also the optimization target of the neural network; Step 2: Change the noise intensity, construct noisy data and corresponding labels under noise interference of different intensities, and form a sound signal denoising dataset to improve the feature extraction and generalization capabilities of the neural network model; Step 3: Construct a two-stage sound signal denoising neural network model CBDNet. The neural network includes a noise estimation subnetwork and a denoising subnetwork. The noise estimation subnetwork is responsible for estimating the noise intensity in the sound signal, including information such as noise amplitude, phase, and signal-to-noise ratio. The denoising subnetwork performs accurate noise removal based on the noise information estimated by the noise estimation subnetwork, and completes the training and testing of the neural network model on the data set constructed in step 2. Step 4: The microphone on the wireless Bluetooth headset collects the noise signal in the surrounding environment. The noise signal processor inside the headset will perform reverse processing on the collected noise signal to generate a sound wave signal with a phase difference of 180 degrees from the noise and approximately the same amplitude. Step 5: The generated sound wave signal opposite to the noise is mixed with the audio signal and noise signal played by the headset and sent to the output system of the headset to achieve partial noise cancellation. Step 6: The mixed sound signal after the cancellation process in Step 5 is sent to the neural network model that has completed training and testing in Step 3 to achieve further refined elimination of the noise signal. Step 7: The sound signal after two noise reduction processes is output to the human ear through the speaker of the headset, thereby reducing the interference of external noise on the user.

5. The intelligent control method of a wireless Bluetooth headset according to claim 1, characterized in that: The passive noise reduction method designs the physical structure of the wireless Bluetooth headset as an in-ear headset. By closely fitting the earplug with the ear canal of the human ear, a relatively closed environment is formed, and external noise is isolated through physical noise reduction methods to reduce noise interference.

6. The intelligent control method of a wireless Bluetooth headset according to claim 1, characterized in that: The call noise reduction method uses the noise reduction chip inside the wireless Bluetooth headset to filter the signal received by the call microphone, reduce interference such as wind noise from the outside world, and at the same time can perform echo attenuation processing on the call sound transmitted by the other party.

7. The intelligent control method of a wireless Bluetooth headset according to claim 1, characterized in that: The hash algorithm is an algorithm, and the specific steps include: Step 1: Padding. Output the length of the input voice signal in bits, and divide this length by 512 to take the remainder. If the remainder is not equal to 448, then one 1 and n 0s need to be filled so that the remainder of the filled information length divided by 512 is 448. At this time, the length of the information is: N * 512 + 448 (bits). Step 2: Record the information length. Use 64 bits to store the information length before padding, and place these 64 bits behind the result of Step 1. At this time, the information length is N * 512 + 448 = (N + 1) (bits). Step 3: Load the standard magic numbers (four integers). The standard magic numbers are all hexadecimal numbers, and the physical order is: A = (01234567)16, B = (89ABCDEF)16, C = (FEDCBA98)16, D = (76543210)16. Step 4: Four rounds of cyclic operations 1) Set the maximum number of loops to the number of groups plus 1. 2) Divide each 512-byte into 16 cyclic groups, each group with 64 bits, that is, 8 bytes. 3) Define four linear functions F(X,Y,Z)=(X&Y)|((~X)&Z) G(X,Y,Z)=(X&Z)|(Y&(~Z)) H(X,Y,Z)=X^Y^Z I(X,Y,Z)=Y^(X|(~Z)) Among them, & represents the "AND" operation in logical operations, | represents the "OR" operation in logical operations, ~ represents the "NOT" operation in logical operations, and ^ represents the "XOR" operation in logical operations. 4) Let Mj represent the jth sub-group of the message (from 0 to 15), and <<<s represent a cyclic left shift of s bits. Then the four-round operation can be defined as: FF(a, b, c, d, Mj, s, ti) means a = b + ((a + F(b, c, d) + Mj + ti) <<< s) GG(a, b, c, d, Mj, s, ti) means a = b + ((a + G(b, c, d) + Mj + ti) <<< s) HH(a, b, c, d, Mj, s, ti) means a = b + ((a + H(b, c, d) + Mj + ti) <<< s) II(a, b, c, d, Mj, s, ti) means a = b + ((a + I(b, c, d) + Mj + ti) <<< s); 5) Four-round operation The first round a = FF(a, b, c, d, M0, 7, 0xd76aa478) b = FF(d, a, b, c, M1, 12, 0xe8c7b756) c = FF(c, d, a, b, M2, 17, 0x242070db) d = FF(b, c, d, a, M3, 22, 0xc1bdceee) a = FF(a, b, c, d, M4, 7, 0xf57c0faf) b = FF(d, a, b, c, M5, 12, 0x4787c62a) c = FF(c, d, a, b, M6, 17, 0xa8304613) d = FF(b, c, d, a, M7, 22, 0xfd469501) a = FF(a, b, c, d, M8, 7, 0x698098d8) b = FF(d, a, b, c, M9, 12, 0x8b44f7af) c = FF(c, d, a, b, M10, 17, 0xffff5bb1) d = FF(b, c, d, a, M11, 22, 0x895cd7be) a = FF(a, b, c, d, M12, 7, 0x6b901122) b = FF(d, a, b, c, M13, 12, 0xfd987193) c = FF(c, d, a, b, M14, 17, 0xa679438e) d = FF(b, c, d, a, M15, 22, 0x49b40821) The second round a = GG(a, b, c, d, M1, 5, 0xf61e2562) b = GG(d, a, b, c, M6, 9, 0xc040b340) c = GG(c, d, a, b, M11, 14, 0x265e5a51) d = GG(b, c, d, a, M0, 20, 0xe9b6c7aa) a = GG(a, b, c, d, M5, 5, 0xd62f105d) b = GG(d, a, b, c, M10, 9, 0x02441453) c = GG(c, d, a, b, M15, 14, 0xd8a1e681) d = GG(b, c, d, a, M4, 20, 0xe7d3fbc8) a = GG(a, b, c, d, M9, 5, 0x21e1cde6) b = GG(d, a, b, c, M14, 9, 0xc33707d6) c = GG(c, d, a, b, M3, 14, 0xf4d50d87) d = GG(b, c, d, a, M8, 20, 0x455a14ed) a = GG(a, b, c, d, M13, 5, 0xa9e3e905) b = GG(d, a, b, c, M2, 9, 0xfcefa3f8) c = GG(c, d, a, b, M7, 14, 0x676f02d9) d = GG(b, c, d, a, M12, 20, 0x8d2a4c8a) The third round a = HH(a, b, c, d, M5, 4, 0xfffa3942) b = HH(d, a, b, c, M8, 11, 0x8771f681) c = HH(c, d, a, b, M11, 16, 0x6d9d6122) d = HH(b, c, d, a, M14, 23, 0xfde5380c) a = HH(a, b, c, d, M1, 4, 0xa4beea44) b = HH(d, a, b, c, M4, 11, 0x4bdecfa9) c = HH(c, d, a, b, M7, 16, 0xf6bb4b60) d = HH(b, c, d, a, M10, 23, 0xbebfbc70) a = HH(a, b, c, d, M13, 4, 0x289b7ec6) b = HH(d, a, b, c, M0, 11, 0xeaa127fa) c = HH(c, d, a, b, M3, 16, 0xd4ef3085) d = HH(b, c, d, a, M6, 23, 0x04881d05) a = HH(a, b, c, d, M9, 4, 0xd9d4d039) b = HH(d, a, b, c, M12, 11, 0xe6db99e5) c = HH(c, d, a, b, M15, 16, 0x1fa27cf8) d = HH(b, c, d, a, M2, 23, 0xc4ac5665) The fourth round a = II(a, b, c, d, M0, 6, 0xf4292244) b = II(d, a, b, c, M7, 10, 0x432aff97) c = II(c, d, a, b, M14, 15, 0xab9423a7) d = II(b, c, d, a, M5, 21, 0xfc93a039) a = II(a, b, c, d, M12, 6, 0x655b59c3) b = II(d, a, b, c, M3, 10, 0x8f0ccc92) c = II(c, d, a, b, M10, 15, 0xffeff47d) d = II(b, c, d, a, M1, 21, 0x85845dd1) a = II(a, b, c, d, M8, 6, 0x6fa87e4f) b = II(d, a, b, c, M15, 10, 0xfe2ce6e0) c = II(c, d, a, b, M6, 15, 0xa3014314) d=II(b,c,d,a,M13,21,0x4e0811a1) a=II(a,b,c,d,M4,6,0xf7537e82) b=II(d,a,b,c,M11,10,0xbd3af235) c=II(c,d,a,b,M2,15,0x2ad7d2bb) d=II(b,c,d,a,M9,21,0xeb86d391); 6) After each cycle, add A, B, C, and D to a, b, c, and d respectively, and then enter the next cycle; 7) The MD5 algorithm performs MD5 verification on the user data sent by the Bluetooth headset and the data received by the mobile terminal to prevent the user data from being maliciously tampered with; 8) The user password in the mobile phone system is stored in the file system after MD5 hash operation. When the user logs in, the system performs MD5 hash operation on the password entered by the user, and then compares it with the MD5 value stored in the file system to determine whether the entered password is correct. Through such steps, the system can determine the legitimacy of the user's login to the system without knowing the plain text of the user's password, thereby protecting the user's personal information security.

8. An intelligent control system for a wireless Bluetooth headset, characterized in that: The intelligent control method of a wireless Bluetooth headset according to any one of claims 1 to 7 is used to realize an intelligent control system of a wireless Bluetooth headset, wherein the intelligent control system of the wireless Bluetooth headset includes an intelligent touch screen, a Bluetooth data communication system, a data encryption transmission link, a voice signal processing system and an intelligent voice noise reduction system: The smart touch screen uses pressure-sensitive touch technology, with a built-in capacitor plate, and the user's finger is used as the other capacitor plate. When the user presses or taps the smart touch screen, the distance between the two capacitor plates changes, and the current on and off state also changes. Different built-in instructions correspond to different user tapping rhythms, and different operation instructions are executed. At the same time, the built-in fingerprint recognition system can intelligently identify the user's identity based on the fingerprint information recorded by the connected mobile phone, ensuring the safety of the user's wireless Bluetooth headset. The Bluetooth data communication system is used to realize data interaction between users, mobile phones and wireless Bluetooth headsets. It adopts Bluetooth 3.0 communication protocol and transmits data through Wi-Fi at a transmission rate greater than 1Mb / s. The modulation method is Gaussian frequency shift keying modulation with a modulation index range of 0.28-0.

35. A Gaussian filter is used to smooth the frequency switching process of the modulated signal, convert the digital signal into a continuous sine wave signal, and transmit information by changing the offset of the carrier frequency, thereby improving the signal's spectrum efficiency, anti-interference ability and anti-frequency deviation ability. The data encryption transmission link uses encryption algorithms to encrypt and protect user voice data, text data and other personal data to avoid external noise, malicious interference and personal privacy leakage; The voice signal processing system consists of two parts: the voice signal acquisition subsystem and the voice signal intelligent recognition subsystem. The voice signal acquisition subsystem contains a single microphone, which is responsible for collecting the user's voice signal; the voice signal intelligent recognition subsystem has a built-in neural network model, which can effectively recognize and understand the user's voice signal and execute related instructions; The intelligent voice noise reduction system consists of a noise signal processor, an intelligent noise reduction subsystem and a noise reduction chip. The noise signal processor is located at the back end of the microphone and analyzes the collected noise signal to provide parameter information such as noise signal power, amplitude, phase, etc. It also generates a sound wave signal with a phase difference of 180 degrees and roughly the same amplitude as the noise to offset some noise interference; the intelligent noise reduction subsystem contains a corresponding neural network model to further refine the noise reduction processing of the sound signal after the initial noise reduction by the noise signal processor; the noise reduction chip inside the wireless Bluetooth headset filters the signal received by the call microphone to reduce external interference such as wind noise. At the same time, it can also perform echo weakening processing on the call sound transmitted by the other party to improve the user's call quality.

Images